Arizona State University
Arizona State University
Abstract. In this paper we argue that the communicative constitution of organizations requires not just one, but four types of messages, or more specifically types of message flow or interaction process. Such a variety of message flows is required because complex organizations require distinct types of relations to four "audiences". They must enunciate and maintain relations to their members through membership negotiation, to themselves as formally controlled entities through self-structuring, to their internal subgroups and processes through activity coordination, and to their colleagues in a society of institutions through institutional positioning. These four sorts of communication are analytically distinct, even though a single message can address more than one constitutive task; we need to recognize that complex organizations exist only in the relatedness of these four types of flow.
Max Weber founded modern organization studies by offering an interpretive analysis of bureaucracy (1922/1968). His account can be summarized as follows: members use the ideal type conception of bureaucracy to understand the conduct of other members and to guide their own actions; because they all act in patterns organized by the ideal type, their actions coordinate in such a way that organizations consequentially and meaningfully exist. Thus, from its beginning, organization studies have pursued the central question of how large-scale, purposefully-controlled organizations are constituted. In this paper, we will attempt to help answer this question by presenting a theoretical framework for the communicative constitution of complex organizations. We will begin with a selective and partial review of important theories that basically argue for a communicative approach to understanding the nature of organizations. Then, we discuss the meaning of the phrase "communicative constitution of organizations" by specifically defining the terms constitution, organization, and communicative. The central focus of the paper will be the presentation of our theoretical framework consisting of four types of constituting communication processes or what we call "flows." The four flows involve the processes of membership negotiation, organizational self-structuring, activity coordination, and institutional positioning. Through the explication of these four flows we will be able to argue for a theoretical framework that takes both micro- and macro-level issues into consideration in analyzing the communicative constitution of organizations.
Weick (1979) brought to the forefront for modern communication theorists the idea that organizations were not mere objects or systems that existed physically. For Weick, organization was the process of organizing, of interpreting an enacted environment in a way that led to orderly action. His theoretical move from organization being a static entity to a dynamic process was a dramatic turn in how organizational communication could be studied and explained. The basic theme for his organizational model can be found in the recipe for sensemaking: "How can I know what I think until I see what I say?" (Weick, 1979, p. 133) This recipe is understood as a combination of three processes: enactment, selection, and retention; patterns of sense-making action and communication reflectively identified and retained by members add up in retrospect to a social entity called "an organization."
Sensemaking occurred as organizations, or at least organizational members, talked to each other and retrospectively made sense of the talk which could then be stored as knowledge for future use (Weick, 1979). While Weick was far from the first theorist to take what we might call "the process turn," his varied and equivocal formulation was tantalizing enough to exert far-reaching influence across many camps of organizational communication research. But Weicks image of organizing allows it to occur even among people, in the minimal social situation, who are unaware of each others existence. We seek an account that focuses specifically on entities more like the complex formal organizations of todays world.
Another theorist, with motives quite different from those of Weick, was Smith (1993), who explicated the relation between communication and organization by identifying root-metaphors that undergirded the discourse of organizational communication. The notion of root-metaphor, while allowing critique of theories, also offers an entry-point into the ontology of organizations. Root-metaphors capture a fundamental, underlying worldview and are able to undergird broad areas of meaning (Smith and Eisenberg, 1987). Smiths work made the organization-communication relation a central problem recognized by the field; and furthermore, suggested that reconceptualiztion of the object/unit of analysis could avoid the problems of reification and marginalization. She argued that organizational communication theories were cast in terms of containment (that organizations involve spatial limits within which communication processes occur), production (that either communication or organization is the produced, even causal outcome of the other process), and equivalence. Because of limitations in the employment of the first two metaphors, she states that many writers have argued for an equivalency positionthe idea that communication is organization and organization is communication. For Smith, the weakness in this position is that, "If organization and communication are equivalent, to explicate Organizational Communication in terms of organization-communication would be tautological" (Smith, 1993, p. 28). While she does not offer an elaborated answer to the question of the communication-organization relation, she sharpened our perception of the importance and neglect of that question and moved the field toward potentially greater rigor. But she also, through the very generality of her metaphors, allowed the impression to persist that the constitution problem can have a basically simple answer.
Boden (1994) approaches the constitution issue by examining how single communication events such as telephone calls, gossip, or planning meetings structure organizations. Her goal was to demonstrate that interactive mechanisms could implicate organizational properties. By applying the technique of conversation analysis to ordinary communication events that occur in organizations, one should be able to observe the interface between talk and social structure. For example, if a researcher studied a planning meeting, she could not only analyze the adjacency-pairs of turn-taking, but examine how people use "turn-making" to advance their own political position (p. 18). However, Bodens argument is weakened by lack of a justified list of necessary and sufficient organizational propertiesshe discusses one organizational phenomenon after another, but does not show or argue that they "add up" to organization. Furthermore, Boden gets her list of phenomena from established organizational theories, not from specifically communication theory. Thus, her argument shares the weakness of many reductionist arguments, that it does not capture the emergent relations among phenomena that are essential to organizational studies. Studying the fragments of conversation that occur between organizational members does offer insight at the micro-level of how talk creates structure; yet this position does not provide an explanation of how all the single communication events synthesize to constitute an entire organization at the macro-level.
Taylor (1993) also tackled the discussion of what organizational communication theory should entail as he states that, "The goal of organizational communication theory ought to be to bridge the micro/macro gap, by showing how to discover the structure in the process and delineating the processes that realize the structure." (p. 261) The processes of communication create a patterning that constitutes the structure of organization and the organization itself simultaneously. To develop his conception of communication, Taylor turns to Greimas (among, we should hasten to note, many other theoretic strands), who contends that all communication has an underlying deep narrative structure that organizes conversation through various speech acts. The constitution of an organization would involve the deep narrative structuring of a great number of elementary transactions conducted by human agents. Another main tenet is his claim that communication involves two aspects, conversation and text, with the latter (the medium of organizational structure) stabilizing and grounding, but also being enacted and potentially transformed by, the former (the medium of organizationally communicative action). Since communication creates the structure of organization, Taylor argues that it makes sense to study organizations from the communication perspective. A key point to his position, which seems to be comparable to Weicks, is that organization is an effect of communication and not its predecessor. Taylor vastly extends the range of communication theory applied to the constitution problem, but his fascination (even as a pronounced interpretivist) with structuralism leads him to root his answer to the constitution problem in a grammatical rather than a systems conception.
Deetz and Mumby (1990) provide an additional view of how organizations are constituted. First they remind us that organizations are not simply given in their current form and persist through time, but they have to be produced and reproduced continually. Second, these organizations also exist within a superstructure over time and space that includes values, laws, rules, ideology, and other institutions indeed, their development represents a quantum leap in social abilities to concentrate and exert power. Finally, although they agree that communication is constitutive of organizational reality, Deetz and Mumby integrate the issue of power into the constituting process. Their position is that "power is an inevitable and constitutive element in all social and institutional interaction . . .. All communication necessarily involves the use of power, and the role of a radical theory of organizational communication is to explicate the processes through which power is manifested and thus shapes organizational reality" (p. 37). Discursive practices that are employed every day by members of organizations aid in the constitution of meanings in their organizational lives. Furthermore, communication is understood to be ideological because it produces and reproduces particular power structures to the exclusion of alternate power configurations. As organizational members engage in communicative practices in certain ways, they are indeed shaping and constituting their organization into an unique formation that is very different than other organizations. Deetz and Mumby bring our attention usefully to the historical, practical, and power-relevant sides to the constitution question, and discuss vital processes that we shall re-describe below. However, their main concern is not to provide a general account of how organizations are constituted, but to argue how central the question of power is to organizations.
In short, we might single out four contributions, tagged simplistically by keyword labels, that past theory has made to the constitution question: the idea of process, the question of equivalence, the idea of structure, and the idea of power. In the theoretic framework presented below, we seek to draw on these ideas while introducing a fifth: the idea of multiplicity.
A definition or explication of terms is very important for a paper on this topic. There are three important ones for us, to be discussed in the following order: constitution, organization, and communicative.
In his magisterial Constitution of Society, Giddens (1984) mentions "the constitution of day-to-day life," context as drawn upon by actors in "constituting communication," and "the phenomenon of talk . . .as constitutively involved in encounters" (pp. xxii, 71, 73). However, he never explicitly defines "constitution," or places it in his index, or gives it sustained focal discussion. Therefore, it might be useful for the sake of guidance to spend a little time explicating the phrase "communicative constitution of organizations." "Constitution" is a technical term in philosophy, especially in Kantian philosophy, Marxism, and phenomenology. It is rooted in the Kantian notion that objects and causal relations have reality only due to the activity of the transcendental ego. "Constitution" has a variety of philosophical meanings ranging between the epistemological (we, as researchers, know reality by constituting concepts of it) and the ontological (we, as members of society, constitute our reality, or social reality), between the cognitive and the practical/active, between creation and making sense of what already exists (Outhwaite, 1983). It also has technical meaning in speech act theory, especially in the distinction between regulative and constitutive rules (Searle, 1969; Giddens, 1984, p. 20). Constitutive rules are those which define an institutionalized speech practice, making it what it is.
All of these schools seemingly influence Giddens, the last most strongly. We would claim that he uses "constitute" so as to point to his concept of the duality of structure. As agents behave, they constitute interaction and its meaningful units because meanings, communicative acts, and episodes are what they are only due to the knowledgeable, empowered, contextually positioned action that implicates them. This reflexive dependence of action and meanings extends to institutions as well: "The fixity of institutional forms does not exist in spite of, or outside, the encounters of everyday life but is implicated in those very encounters" (1984, p. 69). His sense of constitution is primarily ontological and practical, but rather than creation involves reproduction and transformation.
We use roughly Giddens sense of "constitution" below: a pattern or array of types of interaction constitute organizations insofar as they make organizations what they are, and insofar as basic features of the organization are implicated in the system of interaction. This relevance is not necessarily outside the knowledge of members and others who are communicatingwhile they may see themselves as powerless to destroy or fundamentally change the organization, they typically do know how to make their communication compliant to dominant organizational directives, or resistant, or irrelevant and non-organizational. After too many resistant choices by members, the climate of the organization may change and its legitimacy may sink, even in the face of top member resource control. So communication even by members low in power still does forceful work on the constitutive task.
Following McPhee, Corman, and Dooley (1999), we see much to admire in Jelinek and Litterers (1994) definition of an organization as a "deliberately created and maintained social institution within which consciously coordinated behaviors by members aim to produce a limited set of intended outcomes" (p. 12). We do approve the emphasis in this definition on institution-hood, though the idea that all organizations are institutions may exaggerate their fixity and conscious coordination with purposeful intent (however, see Giddens (1976, 1984) critique of these concepts). Nonetheless, we explicitly do not assume that the institutionalized organization is an unquestionable given, that interaction across or outside institutional boundaries is inessential, or that most behavior "inside" the organization, even contributing to its persistence, is either conscious or coordinated. More importantly, this definition implies a model of the organization as behaviors inside an institutionalized container, coordinated by prior plan or cognition. So we would prefer to transform the definition to "a social interaction system, influenced by prevailing economic and legal institutional practices, and including coordinated action and interaction within and across a socially constructed system boundary, manifestly directed toward a privileged set of outcomes."
We can go farther in framing our paper if we explicate the notion of "communicative constitution of organizations" more specifically. We emphasize, first, that all communication has constitutive force. At the very least, it constitutes socially recognized agency: when we communicate, an unstated presupposition accompanying our words is that the speaker is a conscious, capable agent; when hearers interpret our words, they use the presumption of agency to help make sense of our words (Grice, 1987). On the other hand, a listener who ignores our words or rejects their validity partly undermines the establishment of agency, so the whole communication process, rather than any one act or exchange, is the locus of constitution. A second point to note is that, although communication relatively straightforwardly constitutes the agency of the communicating parties and aspects of their relationship, the constitution of outside objects, especially complex organizations, is itself more complex. Two people conversing can no more constitute, say, General Motors or the Redheaded League than two Birchers can conversationally constitute a Communist conspiracy. (Here it is important to be precise: of course two conversants can constitute a conspiracy as a topic or assumption of conversation, but generally not as the type of thing it is claimed to be in fact. Insofar as we seek to explain organizations as complex distantiated systems, the preceding argument holds.) It seems logical to expect that, in order to constitute a complex organization, a complex relation among organizational communication processes is required. Third, it is important to emphasize that not all communication is organizational. For instance, a casual chat between friends certainly makes them into a communication system, coordinates their perspectives to some degree, and even involves some conversational organization. But the friends are usually not "an organization" as a result of having communicated, in the sense relevant to the tradition of organizational communication or the definition of organization stated above.
Fourth, we will suggest relatively broad and abstract ways in which communication, including both single messages and interactive episodes, constitutes organizations. In other words, we will pitch our analysis one or more levels of abstraction above that of Boden (1994). While she showed that conversational processes can help constitute organizations, we want to identify broad but clear types of processes being carried out in the conversation. We doubt that proceeding inductively by identifying scripts or longer recurrent conversational segments will work, given the variety of organizational cultures and ways constitution can be carried out. Instead, we will proceed more deductively, identifying types in terms of their necessity for a complex organization to exist and have the impact it does in society.
For us emphasizing communication means emphasizing circulating systems or fields of messages. We will follow Mintzberg (1979), and more recently Lash and Urry (1994), in calling these "flows," but we emphasize that these flows involve crosscurrents, and are considered as constitutive communication, not merely information transmission. Thus each episode of communication is interactive, involving multiple participants with only partly shared goals and understandings; the results of communication episodes are by no means physically "transmitted," but become conditions mediated in later interaction episodes involving the initial parties or others. As a result of "chains" of interaction episodes, certain topics and ideas become manifest in successively larger domains of the organization (Sperber and Wilson, 1986), but any resemblance of an organizational communication process to an electrical network is the result of a definite array of social practices strategically engaged in by agents, admittedly under institutional and other conditions that bolster the "networkness" of the result.
The need for a larger and more general unit of analysis might be easier to convey metaphorically. Think of an organization as a collection of member cells, with messages as the blood, the hormones, the nerve impulses that affect and relate them. Of course, specific chemicals and nerve-signals affect specific cells. However, once we recognize those effects, another problem remains how do we account for the nature and growth of whole organs or bodies? We need to register the whole array of necessary influences and types of influence on the organ and the pattern of their effects.
Similarly, it is vital to begin by identifying the types of flows that make an organization what it is, and to plot their interrelations. To state the argument less metaphorically than above, organizations are complex and have varied defining facets, so that no one grammatical or communicative form is sufficient to constitute them. On the other hand, they are so varied in size, origin, and member status, and thrive so persistently through changes of membership and structure, that a theory of constitution must be highly general, allowing organizations to occur in a variety of ways. Although specific messages can be decisive in the outcome of a decision-making session, for instance, no specific message or even decision session is necessary or decisive for making the group of members an organization. But decision-making sessions, as a type of message flow or interaction process, might be essential.
Our analysis compares and contrasts with two threads of argument implicit in the theoretical writings of Taylor and his colleagues. In one thread, they argue that specific grammatical forms (for instance, ditransitive forms or more narrowly, commands, or as another instance, narrative form (Taylor, 1993; Taylor, Cooren, Giroux, & Robichaud, 1996) are constitutive of or fundamental to organizations. Usually they argue this by showing how important organizational processes can be represented/enacted, or can only be represented/enacted, by using the grammatical form in question. We see this kind of argument as valuable in pointing out how essential functions of organizing are rooted in communication, and possibly in identifying formal communicative features to focus on in analysis of discursive transcripts. On the other hand, the necessity Taylor and his colleagues face in focusing on broad grammatical features is that the grammatical forms have limited power to distinguish and explain complex social forms such as organizations. For example, commands and narratives occur importantly in marriages and casual chat just as in corporate communication. And in focusing on the "command" form, they may abstract it from a type of discourse flow in which it is an important but incomplete part, like a cell nucleus studied apart from its role within a cell.
A discussion of the ongoing constitution of an enduring systemic form such as an organization automatically raises the issue of functionalism. Are we presuming the existence, stability, orderliness, universal utility, and even self-sustaining powers of organizational systems in a way which derogates the agency of human individuals or their unequal power and treatment? Nowe explicitly deny these assumptions. But we believe that a more limited version of functionalism is unavoidable or at least useful in discussing the topic of the persistence of organizations and societies of organizations. Organizations are a social form created and maintained by manifestly and reflexively reifying practices of membersthe members think of, treat, and relate to organizations as real, higher-order systems, and make provision for their survival. In addition, some communication patterns may contribute to the existence and persistence of organizations as an unintended consequence, and may even be necessary for their survival. Any analysis that points this out will sound functionalistic; we do not presume that the commonsense existence of organizations is real in any sense beyond its reality within and conditioning of the practices of members, or that such "reality" is free from ambiguity, aporias, or contradictions. But interpretive, postmodern, and even some critical analyses are sometimes phrased so as to imply that organizations are unreal figmentsthat "General Motors" is not a thing that could be causally relevant to lives of hundreds of thousands of people or the existence of automobiles on roads. This implication, we think, is silly. More scarily, the form of our theory is eerily suggestive of Parsons (Parsons and Smelser, 1965) four-function scheme that dominated 1950s sociology and stimulated the currently fashionable ire against functionalism. His models acronym is AGIL, which represents the four basic functions necessary for the persistence of a social system such as an organization. The letters of the acronym stand for Adaptationthe problem of acquiring and using resources; Goal Attainmentthe problem of setting, legitimizing, and implementing goals relative to higher-order systems; Integrationthe problem of maintaining solidarity or coordination among subsystems; and Latencythe problem of creating, preserving, and transmitting the systems distinctive culture and values. (See Rollag, n. d.) This approach was criticized for conservatism, in underplaying the role of contradiction and change in systems, and for mechanism, in using an oversimplified model of agency. In contrast, our analysis will expose the critical and interpretive edges of organizational constitution.
For instance, we see this theory as having three values. First, it sketches an explanation of the power and efficacy of organizations in the West-dominated world today. They are the kind of thing that can have such power because they constitute themselves in the four ways noted below. As Perrow indicates (1979), such theories have critical import. Second, its four flows of messages are actually more or less hidden implications of conversations and reports within and outside organizations, operating on a level that may not be obvious or seem important to members. Explicating such implications and presuppositions is a hermeneutic task, potentially allowing members to understand their own communication better. Third, these flows are arenas in which organizations do vary and can be changed in their fundamental nature. Many authors have claimed, over the decades, that new forms of organizations have emerged, as a result of various social and technological developments. A theory such as this one gives us a template by which to detect, diagnose, and assess novel organizational phenomena.
Before beginning the specific description of the four constitutive communication processes, a brief overview might be useful. We want to argue that organizations are constituted in four different communicative flows, not just one, and that the flows are different in their main direction and in their contribution to organizational constitution, with each making a different and important contribution. Furthermore, we argue that organizations and communication are varied enough so that we cannot go much further in explaining constitution at this level of generality than by discussing types of flow. We see our theory as building on and elaborating the theoretical underpinnings reviewed above, summarized by the four keywords of process, equivalence, structure, and power. Our emphasis on communicative flow takes up Weicks idea of process; our four flows escape tautology in showing the equivalence of communication and organization. In each flow, a sort of social structure is generated through interaction; and by allowing for one flow to control or condition another, the model allows for specifically organizational power. The types of flow are analytically differentwhile they are often distinct, a single message can and often does make more than one type of contribution. Also, as mentioned above, each kind of "flow" is actually a kind of interactive communication episode, usually amounting to multi-way conversation or text passage, typically involving reproduction of as well as resistance to the rules and resources of the organization. The four flows link the organization to its members (membership negotiation), to itself reflexively (self-structuring), to the environment (institutional positioning); the fourth is used to adapt interdependent activity to specific work situations and problems (activity coordination). Figure 1 gives a schematic of an organizational system and the four directions of flow.
Organizations always must include members and are distinct in nature from the members. Thus, one process vital to an organization is the communication that establishes and maintains or transforms its relationship with each of its members. We should emphasize the obvious"membership" in any one organization is not a natural property of people, and is instead constituted by/in this flow of communication. But in constituting members, the communication process importantly constitutes the organization, since one must be a member of something.
One of the best-known examples of member constitution is member recruitment and socialization (Jablin, 1987). Prospective members must be evaluated and categorized; both the new member and the organization must decide to create a relationship; and the new member must be incorporated into the routines and structures of the organization, and vice versa. However, in the course of this socialization process, the organization is simultaneously framed as having prior existence, a multitude of other members, and the power to induce a relation of co-membership (as well as other relations like supervision and mentorship) between members. This facet of the individual-organization relationship is well recognized; although some others are less so.
One other facet is the shaping of the member relation itself. What does it mean for Al to be a member of or related to organization O, and how is the answer to that question worked out? A fairly good example is embedded in Delanys new-wave science fiction novel Dhalgren (Delany, 1974). His amnesiac hero, the Kid, occupies an anomalous position on the outskirts of a gang in an anomic future world. The Kid is accepted by the gang, even has some leadership status and respect, yet both he and the gang are constantly aware that he is not a member. Similar membership issues are faced by engineers or managers "loaned" from one organization to work in another. This facet shades over quickly into a second relationship, involving identification and identity (Tompkins and Cheney, 1985).
Interaction processes that might be expected given the nature of this flow include, first, a dialectic of reputation and courtship, including all the varying strategies exhibited during job-seeking and recruitment. It is common knowledge that both the organization and the person typically take the most positive line possible, often tacitly offering to redefine themselves to fit the others expectations more closely. A second web of processes involves identification or positioning by individuals, and inclusion by organizations: these terms are classically used to refer to the problem of membership construction. Finally, we must not forget that the problem of relation between individual and organization exists even for members very high in status; power-claiming and spokesmanship are processes of negotiation of relations of power over organizational resources or the whole organization. For instance, Pacanowsky (1987) notes how Bill and then Bob Gore are clearly in charge of Gore, Inc., yet avoid many labels and rituals common in other companies that would coalesce their power as formal position. Instead, they try to manage an ambiguous role as equal, yet inevitably "more equal," organization member.
Why is this process a vital facet of communicative constitution of organizations? One answer is that organizations, like all social forms, exist only as a result of human agency (Giddens, 1984). By many definitions of communication, only individual humans can communicate, so when communication constitutes organization, the relation of the communicators to the organization is important. We would want to go farther, though, to emphasize that organizations ineluctably involve members, almost, as the metaphor suggests, as parts or limbs of the organization. Organizations exist when they draw members in, lead them to take part in and understand the interactional world unique to the organization.
Organizations do not draw members and coordinate work automatically or as a result of natural tendency; some individual or group typically works hard to bring the organization into being, make decisions about such matters as member time and resource investment. In short, organizations are the objects not merely of reflexive attention but of reflexive control and designof self-structuring. We would claim that this reflexive self-structuring distinguishes organizations from groupings such as lynch mobs or mere neighborhoods; it is essential to the explanation of the power of formal organizations in history, especially but certainly not only Western economic history (McPhee, 1985). It is important to emphasize that self-structuring is a communication process among organizational role-holders and groups; it is analytically distinct from, though often part of the same messages as, communication that helps coordinate the activities of members. It is unique in that it does not directly concern work, but rather the internal relations, norms, and social entities that are the skeleton for connection, flexing, and shaping of work processes.
Examples of communication like this are easy to giveif anything, they are stereotypical of organizational communication. Official documents such as charters, organization charts, policy and procedure manuals; decision-making and planning forums; orders, directives, and the more casual announcements that often substitute for them; processes of employee evaluation and feedback; budgeting, accounting, and other formalized control processesall these are mainly media for organizational self-structuring. Self-structuring communication includes any process that serves to steer the organization or part of it. It also involves processes that design the organization, the setting up of subsystems, hierarchical relationships, and structural information-processing arrangements (Galbraith, 1973). Johnson (1981, chap. 7) gives examples of how recursively evolving and dialogic communication is responsible for creating and documenting formal structure. Examination of several borderline cases reveals the impact and uses of self-structuring communication. Larson (1992) discusses the process of network organization construction among entrepreneurial firms. In the cases she examined, contracts and other legal self-structuring mechanisms were not present or at least emphasized. In their place she found an extended process of mutual exploration based on reputation and early cooperation followed by trust building and expectation clarification that laid the groundwork for operational and strategic integration of plans and knowledge stocks. The self-structuring process for the network had to go beyond mere considerations of economic advantage to achieve low uncertainty, high mutual knowledge, and high goal alignment before full cooperation could be risked. If we decide to call Larsons sort of network dyad an organization, we do so partly because of this meta-layer of self-structuring process that grounds and solidifies the collaboration. Another seemingly borderline example is Pacanowskys (1987) discussion of Gore, Inc. Gore operates while avoiding reliance on hierarchy, formal structure, and rigid controls. However, even Pacanowskys account recognizes how senior organization members engage in sustained reflection on organizational operations and stimulate extensive communication about work decisions, to allow for widespread responsibility over work operations rather than centralized responsibility. Self-structuring is made globally collective rather than centralized. In addition, Pacanowsky notes in passing that this diffused responsibility does not eliminate processes that control and monitor the division of labor. In short, there are many forms of self-structuring, but the process itself is vital.
Why is it vital? McPhee (1985) argues that the communication of formal structureone form of self-structuring, though a narrower concept than ourshas two important impacts within organizations. First, it substitutes for what we have called collaborative communication above, by pre-fixing work arrangements and norms rather than let them emerge during collaboration. And next, it is authoritative metacommunication that guides but also controls the collaboration and membership-negotiation processes, that takes organizational processes as an object so that the organization as a whole can deal with its environment and be exploited by powerful interests. In the process of achieving these outcomes, the organization is inevitably internally differentiated or distantiated. In addition to these effects of formalized structuring, we would point to several others as not merely useful (for organizational survival or profit) but constitutive. It is in the process of self-structuring that the organization as a system takes control of and influences itself, not merely to handle immediate problems but to set a persistent routine procedure for response. Only through developing this analogue to a sense of self can an organization avoid problems of over-adaptation, incoherence, and confusion. And this kind of reflexive communication constitutes the organization for itself, a basic process in its overall constitution.
However, by taking self-structuring to be a communication process, we avoid the illusion that it itself is unidirectional, internally coherent, or successful by definition. Self-structuring communication is subject to discrepancy, dispersal, and ambiguity, with varying consequences for the system, subsystems, individuals, and outside interests. It is an interpretive and political process, stuck in socioeconomic traditions that, in the West, favor corporate bureaucracy.
Nonetheless, to contribute even at a minimum to the constitution of the organization, communicative interchanges of this sort must assume and implicate a sense of the organization as a differentiated yet purposeful whole. In relatively complex organizations, communication usually must tacitly recognize a governance structure with legitimate power, and whether in implementing, serving, subverting, or resisting it, must reproduce it.
Organizations, by definition, have at least one manifest purpose, and the activity of members and subgroups is partly directed toward it. To a substantial extent, these activities are coordinated as a result of the organizations self-structuring, which creates a division of labor, a standard task-flow sequence, and a series of policies and plans for work. However, such structural directions can never be complete or completely relevant, are never completely understood, and are frequently amended in an informal patchwork of adjustments. In addition, exceptions and problems arise frequently and require coordinated adjustments out of the ordinary (Perrow, 1967). The process of adjusting the work process and solving immediate practical problems requires the sort of communication we call activity coordination.
The clearest exponent of activity coordination as a vital organization-constituting process is probably Barnard (1938), who presents organizations as cooperative systems. All the activities of executives are dependent, in Barnards view, on the cooperative work they support and the cooperative stance taken by workers toward executive arrangements. More recently, developers of systems of computer-assisted cooperative work have devoted tremendous effort to the analysis of cooperative work interaction. For instance, Filippi and Theureau (1993) studied work in a control room for the Paris metro, uncovering a complex weave of mutual assistance and attentiveness that resolved train system breakdowns. They found some general principles of coordination. For instance, the Controller "who starts handling a disruption is responsible for it during its entire course, because he knows all the surrounding circumstances and the consequences of his own decisions." As well, all Controllers "actively listen ...to the details of the solving of an incident...to be able to anticipate delays and amendments...on their won sector" (1993, pp. 182, 181). Of course, the real challenge comes when circumstances force principles like these to be violated.
Exemplary accounts of coordination are well developed by structural contingency theory (Mintzberg, 1979). In Mintzbergs description there are five (later seven) kinds of coordination processes. The most obviously relevant example is mutual adjustment, with highly skilled or mostly unorganized members working out solutions to problems on the spot. Nevertheless his other processes also involve activity coordination, as workers, say, determine how to substitute for one another on an assembly line and how to relieve the pressures of line work.
In activity coordination, as in the other flows, one finds multiple processes and attitudes toward the organization. For example, members can coordinate on how not to do work, or coordination may be in abeyance as members seek power over one another or external advantage for themselves from the system. Nonetheless, what seems inescapable is that members presume that they are working not just on related tasks but within a common social unit with an existence that goes beyond the work interdependence itself. This presumption may be a result of self-structuring discourse.
Institutional Positioning in the Social Order of Institutions.
One other type of communication flow remains to be discussed, communication outside the organization, to other entities, "at the macro level" in systems or functional terms. Such entities include suppliers, customers, and competitors and collaborators, including merger or acquisition candidates. Probably also more powerful organizations such as potential buyers and governmental regulators could be added. Sometimes this communication is presented as a direct product of the focal organization to which it is responsible as a formal entity. More often the communicators are individuals on boundary-spanning roles who negotiate terms of recognition of the organizations existence and place at the same time as they negotiate their own relationships. "Identity negotiation" is an appealing label for this type of communication; we have chosen the broader term "positioning" because the latter includes both identity establishment and development and maintenance of a "place" in the inter-organizational or larger social system. Since identity is inescapably comparative and relational, these two processes merge inextricably.
For example, much of the work of the "institutional school" of organizational theory contains examples of this sort of communication, since an "institution" is recognized as such by and within a community of its peer and related organizationsits "organization-set" in Evans words (1966). Thus, Meyer and Rowan (1977) note how a formal organization chart and similar documents are valuable in establishing the image of legitimacy and rationality presented to fund-suppliers and other peer institutions in the community. The whole process of relationship building involved in capital acquisition is an example of this sort of communication flow. Another broad array of examples is surveyed by DiMaggio and Powells (1983) article on the "iron cage" of institutional isomorphism. The forces bringing about isomorphism are all direct or indirect communication processes, and many of them have force because they create the conditions for future communicative relations. In communication, Cheneys analysis (1991) of the National Council of Catholic Bishops pastoral letter on disarmament illustrates the concern to address other organizations, including other branches of the Church, the Papal See, and both Catholic and general popular audiences.
Of course, all sorts of "business" are transacted between an organization and other agents in its environment, but in the process a number of constitutive moves are required to establish any organization as a "presence" in the inter-systemic institutional order. Thus, whether an organization sells a line of merchandise, attracts capital or donations, or certifies that it has met governmental standards, several processes seem almost unavoidable. The focal organization must actually connect with and induce return communication with important elements of its environment, and vice versa. It must establish or negotiate an image as a viable relational partnercustomer, supplier, neighbor, for example. This image minimally implies creating an impression that the organization meets the acceptability conditions set up by the government and other vital stakeholders. Classically, Apple Computers gained respect only when its managers negotiated an image that allowed them to secure capital and marketing access; start-up companies are often marginal because they lack such reassuring features as institutional status (that protects their property), routine practices, and relations to suppliers and customers. Illegal organizations like the Mafia are also marginal, though even they depend on relations to dummy corporations and use force partly because they lack access to some usual economic institutions. More secure organizations build relationships of trust with important others, and even try to gain control of the uncertainty in their environments in various ways (Pfeffer & Salancik, 1978).
This sort of communication is vital for constituting organizations because organizations exist in human societies that already are organized, that already have institutional ways of maintaining order, allocating material resources, regulating trade, and dividing laborand, of course, that already have ways of communicating about all these practices. Without an institutional backdrop, any but the most primitive human organization is unthinkable; certainly todays complex organizations depend on political, cultural, economic, social, and communicative institutions for their constitution. If each new organization had to reinvent the concepts of property rights and contracts, membership and management, as well as the kind of organization they are (corporation, social service agency), there would be few organizations in existence today. Moreover, institutions like these exist partly because they allow inter-organizational relationsthey allow each organization to draw on other organizations for the variety of resources that it needs to accomplish its goals and maintain itself. Whether or not a completely autonomous organization could exist, in practice, most depend on others, and so, in this dependence, organizations must constitute themselves as practical relational partners.
However, as the proliferation of new organizational forms suggests, there is no one configuration that an organization must use to present its identity to the institutional community. The minimum necessary process seems to be negotiating inclusion, and the measure of inclusion is a purely practical one.
We hope the accounts of the four flows above convey both their relatedness and their difference.
The four flows allow four divergent descriptions of organizational processes: the first recounts the struggle of individuals to master or influence their member roles, statuses, and relations to the organization. The second articulates how organizational leaders design, implement, and suffer problems with decision and control mechanisms. The third focuses on members engaging in interdependent work or deviating from pure collaborative engagement. The fourth describes the organization as a partner, often anthropomorphized, in exchange and other social relations with other organizations.
We definitely do not mean to say that if the four kinds of flows exist, an organization automatically has been constituted and exists. To illustrate the problem, we might consider a neighborhood bar. Suppose one evening two bar patrons discuss going out tomorrow to clean up the neighborhood; in another corner, out of earshot of the first, two other patrons discuss the rules that ought to be followed in clean-up efforts (not too early, etc.); and so on for other groups of patrons. Even if all four types of conversation took place, would the neighborhood be constituted as an organization? We would say not. The four flows would need to be more interrelated, more mutually influential.
For one thing, the four flows need to develop and share a realm of mutual topical relevance, within which the relevance of the other flows themselves is also recognized. This sphere of mutual relevance is what we might call organizational knowledge (McPhee, Corman, and Dooley, 1999). One other requirement would seem to be that the legitimate authority of self-structuring, relative to the other flows, is recognized in the other flows. We do not have reason to believe that this set of relations is sufficient, but we cannot think of others. At any rate, we hope it is clear that all of these flows are required, and that a constituted organization is not just a set of flows, but a complex relationship of them.
To go back to the four keywords we used earlier to reference past theoriesprocess, equivalence, structure, and powerwe agree with the arguments that organization is rooted in or painted on the communication process (Taylor, 1993), though for us the relation is emergence of a higher order system. In a sense this is equivalence, and in a sense not. Organization is not simply communication, but a relationship among distinct types of analytically separable processes, so saying that it "is communication" is misleading, especially from the point of view of level of analysis. Among our communication processes are some that are purposefully designed to generate, and more likely than others do generate organizational structure. This requirement is partly due to the inescapable need for structure in organizations, and partly due to a corporatist and systemic ideology rampant in the world today. But perhaps a deeper sense of structure, on a level with Taylors text-conversation dyad, as it specifically applies to organizations, is the structured relation among the four flows/crosscurrents of organizational communication that inform, enable, and constrain one another. Finally, in various ways, these four currents, especially self-structuring, depend on and generate power imbalance and communicative distortions. Yet they ground a unique approach to critical theory, and are not simply reducible to general theories of ideology, distorted communication, power-knowledge relations, or self-practice relations.
We would claim that the practical and theoretical implications of our model are broad but difficult to trace because they could be developed in many ways. For instance, Child (1980) argues that a number of problems of decision-making and coordination, which for us would be "activity coordination," as well as motivation, which might be included in "membership negotiation" in our model," actually stem from inappropriate organizational structure (or "self-structuring"). If problems apparent in one flow can really develop due to patterns in another flow, we clearly must be aware of all four flows and their potential effects when we diagnose organizational problems. On the other hand, problem-solving efforts must also recognize that problems arising in one flow might nevertheless be capable of solution mainly within another flow, as when problems of member resistance to perceived exploitation are alleviated by creation of a structure wherein members can be promoted frequently (Edwards, 1979). Of course, the same points made about practical diagnosis and remedy must apply to theory: theories focused on one of the flows may be blind to implications of other flows for them. For instance, consider attempts to theorize formal organizational structure (McPhee, 1985; Smith, 1990). Students of information technology and new organizational forms have frequently noted that in some important ways, intra- and inter-organizational cooperation (i.e., activity coordination and institutional positioning) based on informal relations and trust have come to supplant formal structure as modalities of organizational control (Morton, 1991). Whether or not these developments alter the theories of formal structure, clearly the latter must take the former into serious account.
We have argued that organizations are constituted in four different communication flows, not just one; that the flows are different in their main direction and in their contribution to organizational constitution, with each making a different and important contribution. Membership negotiation, organizational self-structuring, activity coordination, and institutional positioning underpin the theoretical framework that was presented in this paper. We have argued that such a variety of message flows is required because complex organizations require distinct types of relations to four "audiences." They must enunciate and maintain relations to their members through membership negotiation, to themselves as formally controlled entities through self-structuring, to their internal subgroups and processes through activity coordination, and to their colleagues in a society of institutions through institutional positioning. These four sorts of communication are analytically distinct, even though a single message can address more than one constitutive task. Finally, we hope that this theoretical framework provides a unique explanation of the complex ways in which organizations are constituted communicatively.
Barnard, C. (1938). The functions of the executive. Cambridge, MA: Harvard University Press.
Boden, D. (1994). The business of talk: Organizations in action. Cambridge, UK: Polity Press.
Cheney, G. (1991). Rhetoric in an organizational society: Managing multiple identities. Columbia, SC: University of South Carolina Press.
Child, J. (1980). Organization. London: Harper & Row.
Deetz, S., & Mumby, D. (1990). Power, discourse, and the workplace: Reclaiming the critical tradition. In J. A. Anderson, (Ed.), Communication yearbook, 13 (pp. 18-47). London: Sage.
Delany, S. (1974). Dhalgren. New York: Bantam.
DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48: 147-160.
Edwards, R. (1979). Contested terrain: The transformation of the workplace in the twentieth century. London: Heinemann.
Evan, W. M. (1966). The organization set:: Toward a theory of interorganizational relations. In J. D. Thompson (Ed.), Approaches to organizational design (pp. 173-188). Pittsburgh: University of Pittsburgh Press.
Filippi, G., & Theureau, J. (1993). Analyzing cooperative work in an urban traffic control room for the design of a coordination support system. In G. de Michelis, C. Simone, & K. Schmidt (Eds.), Proceedings of the third European conference on computer-supported cooperative work (pp. 171 - 186). Boston: Kluwer Academic Publishers.
Galbraith, J. (1973). Designing complex organizations. Reading, MA: Addison-Wesley.
Giddens, A. (1976). New rules of sociological method. New York: Basic Books.
Giddens, A. (1984). The constitution of society: Outline of the theory of structuration. Berkeley, CA: University of California Press.
Grice, P. (1987). Studies in the way of words. Cambridge, MA: Harvard University Press.
Jablin, F. M. (1987). Organizational entry, assimilation, and exit. In F. M. Jablin, L. L. Putnam, K. H. Roberts, & L. W. Porter (Eds.), Handbook of organizational communication: An Interdisciplinary Perspective (pp. 679-740). Newbury Park, CA: Sage.
Jelinek, M., & Litterer, J. A. (1994). Toward a cognitive theory of organizations. Advances in Managerial Cognition and Organizational Information Processing, Vol. 5 (pp. 3-41). Greenwich, CT: JAI Press.
Johnson, B. M. (1981). Communication: The process of organizing. Boston, MA: American Press.
Larson, A. (1992). Network dyads in entrepreneurial settings: A study of the governance of exchange relationships. Administrative Science Quarterly, 37: 76-104.
Lash, S., & Urry, J. (1994). Economies of signs and space. London: Sage.
McPhee, R. D. (1985). Formal structure and organizational communication. In R. D. McPhee & P. K. Tompkins, (Eds.), Organizational communication: Traditional themes and new directions (pp. 149-178). Beverly Hills, CA: Sage.
McPhee, R. D., S. R. Corman, & K. J. Dooley (1999, May). Theoretical and methodological axioms for the study of organizational knowledge and communication. Paper presented at the annual convention of the International Communication Association Convention, Chicago.
Meyer, J. W. & B. Rowan (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83: 340-363.
Mintzberg, H. (1979). The structuring of organizations: A synthesis of the research. Englewood Cliffs, NJ, Prentice-Hall, Inc.
Morton, M. S. S. (1991). The corporation of the 1990s: Information technology and organizational transformation. New York: Oxford University Press.
Outhwaite, W. (1983). Concept formation in social science. London: Routledge and Kegan Paul.
Pacanowsky, M. (1987). Communication in the empowering organization. In J. Anderson (Ed.), Communication yearbook, 11. Beverly Hills, CA: Sage.
Parsons, T., & N. J. Smelser (1965). Economy and society. New York: Free Press.
Perrow, C. (1967). A framework for the comparative analysis of organizations. American Sociological Review, 32:194-208.
Perrow, C. (1979). Complex organizations: A critical essay. Dallas, TX: Scott, Foresman.
Pfeffer, J. and Salancik, G. (1978). The external control of organizations: A resource dependence perspective. New York: Harper and Row.
Rollag, K. (n. d.) "Parsons social system (Structural functionalists)," Encyclopedia of organizational theory [Online]. Available: http://www.stanford.edu/~krollag/org_site/encyclop/parsons.html
Searle, J. R. (1969). Speech acts: An essay in the philosophy of language. Cambridge, UK: Cambridge University Press.
Smith, D. (1990). Texts, facts, and femininity: Exploring the relations of ruling. New York: Routledge.
Smith, R. C. (1993). Images of organizational communication: Root-metaphors of the organization-communication relation. Paper presented to the Organizational Communication Division at the International Communication Association Convention, Washington, D. C.
Smith, R., & Eisenberg, E. (1987). Conflict at Disneyland: A root-metaphor analysis. Communication Monographs 54: 367-380.
Sperber, D., & Wilson, D. (1986 ). Relevance. Cambridge, MA: Harvard University Press.
Taylor, J. R. (1993). Rethinking the theory of organizational communication: How to read an organization. Norwood, NJ: Ablex.
Taylor, J. R., Cooren, F., Giroux, H., & Robichaud, D. (1996). Are organization and communication equivalent? Paper presented at a Conference on Organizational Communication and Change: Challenges in the Next Century, Austin, TX.
Tompkins, P. K., & Cheney, G. (1985). Communication and unobtrusive control in contemporary organizations. In R. D. McPhee & P. K. Tompkins (Eds.), Organizational communication: Traditional themes and new directions (pp. 179-210). Beverly Hills, CA: Sage.
Weber, M. (1968). Economy and society. Berkeley: University of California Press. (Original work published 1922)
Weick, K. (1979). The social psychology of organizing (2nd ed.). Reading, MA: Addison-Wesley.
Copyright 2000 Communication Institute for Online Scholarship, Inc.
This file may not be publicly distributed or reproduced without written permission of the Communication Institute for Online Scholarship, P.O. Box 57, Rotterdam Jct., NY 12150 USA (phone: 518-887-2443).