The Nature of Organizational Communication
"In any exhaustive theory of organization, communication would occupy a central place."
Chester I. Barnard
"All human action takes place in a cross-fire of information."
"Communication is a good deal more talked about than understood."
The purpose of this book is to familiarize the reader with the main concepts, viewpoints, and research findings and applications in the field of organizational communication. Our focus in this book is on the ways in which organizational structure affects communication behavior, and vice. versa.
We discuss, for example, how structure can restrict communication flows, leading to problems of distortion and omission, and how solutions to these difficulties can in turn lead to information overload. The existence of informal communication behavior, typified by rumors, and of informal communication roles, such as liaisons and gatekeepers in communication networks, suggests that the formal structure in an organization far from completely determines communication behavior. Furthermore, such approaches as "office landscaping" imply that communication behavior can occasionally determine organizational structure. Instead of viewing an organization as a completely stable structure, we show how external communication across its boundary with the environment is essential to its functioning, and especially to the innovation process. Our use of open system theory directs us to emphasize information exchange with the organization's environment, as well as communication flows within the organization.
So this book is about communication, but about a rather special kind of communication, viz., that occurring in highly structured settings. Generally, communication scholars have avoided studying the way in which structure affects human interaction. Perusal of communication research literature, at least prior to the 1970s, would almost lead one to assume that social structure does not affect human communication. For example, communication theorists who postulated the S-M-C-R (Source-Message-Channel-Receiver) model and similar models of communication did not accord much importance to the nature of the social relationships between source and receiver. Perhaps this shortcoming stems from the largely psychological backgrounds of early communication scientists, who emphasized intraindividual aspects of human communication in their choice of concepts, units of analysis, and paradigms. In any case, most communication research has been conducted in a way that artificially "destructures" human behavior. The present volume seeks to correct this bias by summarizing what is presently known about communication in organizations.
The research approach of communication scientists in the past has seriously underestimated the impact of social structure on communication behavior. Social structure has often been regarded as an extraneous, bothersome influence in studies of communication behavior, and structural variables have simply been ignored. For example, in most laboratory experiments relative strangers are brought together in a transitory and artificial setting for a brief encounter. The full impact of the social relationships among participants in more real-life communication exchanges is hardly replicated. The effects of social structure on communication that were observed in laboratory studies need to be tested in organizational settings before they can be accepted as appropriate principles of organizational communication (Chapter 5).
In survey research on communication, the role of structure is usually depreciated by the research methods used. The individual is usually the unit of response and is often the unit of analysis (Coleman 1958). Such an atomistic approach ignores the relational nature of human behavior. Most communication is reciprocal and transactional, not a unidirectional flow, as most overly simplified models of human communication would seem to imply (Rogers and Bhowmik 1971).
Organization scholars, for their part, have recognized the crucial significance of communication in understanding organizational behavior, but they are more expert in studying organizational structure itself than in understanding the effects of this structure on communication (or vice versa). The intellectual separateness of communication and organization scientists has discouraged the development of organizational communication as an integrated mode of inquiry.
Our purpose here is to accord organizational communication its rightful place as a serious field of study after many years of "stepchild" treatment by communication, as well as organization, scholars. Our assumption throughout this work is that the behavior of individuals in organizations is best understood prom a communication point of view.
Communication behavior has been investigated in many kinds of organizations, ranging from industrial plants to government agencies to universities and military units, and we shall draw on research findings from these various studies in the present book. We also present a series of illustrative readings about communication behavior in organizations. The following reading shows how inadequate communication between officials and resident families helped contribute to the failure of a showcase housing project. The Pruitt-Igoe project constituted an organization by the definition we present in a following section: it was a stable system of individuals working together through a hierarchy of ranks and division of labor, and with common goals. We feel it is useful to analyze the following case from the viewpoint of communication in an organization.
THE RISE AND FALL OF THE PRUITT-IGOE PROJECT
Public housing for the poor is one of the most controversial social issues in the United States, ranking in the same category as racial busing of school children and welfare programs. Nowhere is the controversy better illustrated than in St. Louis's massive Pruitt-Igoe public housing project.
The project was initially conceived in the 1950s as a means of removing the poor blacks of St. Louis from their vicious cycle of unemployment, crime, poverty, disease, and rundown housing. The federal government gave $20 million toward the $36 million cost of providing the finest low-rent housing that money could buy. Minoru Yamasaki, one of America's most promising architects, was employed to design the massive project of thirty-three ,apartment buildings, each eleven stories high. Pruitt-Igoe (named after a black war hero and a local congressman) was hailed as the most progressive poor people's housing project in the United States.
Yamasaki designed the project as a series of tall brick apartment buildings with 200 feet of open space between each of them, in which children could play and adults could stroll. Each building had a series of elevators that were designed to stop only at the fourth, seventh, and tenth stories, where long galleries or walkways were located. The architect expected that these galleries would be used as community centers, for lounging and socializing and for children to play inside.
A manager/mayor was appointed who had been an army psychologist. Pruitt-Igoe was ready for business.
In October 1954, the first family, a Mr. and Mrs. Green and their four children, took up their residence. Monthly rental was only $37. Mr. Green told a newspaper reporter: "Pruitt Homes will be a swell place to live." More of the 12,000 residents began to move in. Each apartment came equipped with electric refrigerator, gas stove, and lavatory, "luxuries" unavailable to most of the new residents in their previous dwellings.
The central assumption of Pruitt-Igoe was that if an individual's environment were changed from slum conditions to a new, modern apartment complex, the individual's behavior would change accordingly.
Architect Yamasaki began to win lavish praise for his design. Architectural Forum predicted that Pruitt-Igoe would change the public housing pattern in all other cities. The skip-floor elevators and the galleries were lauded as creating "vertical neighborhoods." In the face of such praise, it was easy to forget that the residents had not been consulted at any stage in the architectural planning.
Actually, some residents, especially women, were beginning to see the galleries as "gauntlets" through which they had to pass from the elevator to their apartment door. Housewives had to survive a torrent of abuse, spitting, touching, teasing, and taunting. It was an intimidating experience for the toughest of women.
The elevators started to become very dirty. A lack of public bathrooms on the ground floor prompted hundreds of children to have "accidents" in the elevators. Youths would follow old men onto the elevators and rob and beat them, leaving their bleeding victims inside.
Rent collection became a dangerous calling, and so rents went uncollected. By 1960, St. Louis department stores refused to deliver purchases to Pruitt-Igoe residents. Western Union also refused to deliver telegrams in the project, the only address in America to be so treated. City policemen were ordered not to use their sirens in Pruitt-Igoe out of fear of a violent reaction. The open spaces between tile buildings began to fill up with broken glass and abandoned cars. Bands of razor.wielding youths roamed the area. Delivery men were fatal victims; nurses and welfare workers were raped and brutalized. Hundreds of windows were smashed, but city workmen dared not venture to repair them. During one typical 180-day period in 1968, the project had 2 murders, 10 rapes, 50 robberies, 45 cases of aggravated assault, and 103 burglaries.
Junkies tore the Copper sheeting from the building roofs to sell it to scrap dealers in order to support their drug habit. In the winter of 1971, rain from the leaking roofs froze the pipes, which then burst, leaving hundreds of apartments inches deep in ice. The stairwells became so slippery that residents had to tie ropes around their waists to keep from falling down the stairs.
Under these conditions, more and more of the project families began moving out. As more units became vacant, and the remaining families seldom paid their rent, project operating funds became very tight. The city administration responded by raising the rent. Rents were raised twice in 1968, and a third increase was announced for early 1969. The residents rebelled, as many of them found it difficult to pay tile new rent on their low incomes. They began a rent strike that lasted nine months, toppled the city housing administration, and all but bankrupted the public housing authority. But the new administrators were as far removed from the people m Pruitt-Igoe as those they replaced, and equally unable to cope with the project's problems.
Finally, in 1974, the city government reluctantly announced that it was calling in the heavy iron ball and the dynamite charge. The first two buildings were brought down in only ten seconds with hundreds of pounds of gelignite charges. By the end of the year, all thirty-three buildings had been destroyed. And the only remaining residents of the Pruitt-Igoe public housing project were the thousands of rats who flourished in the piles of bricks and rubble.
Instead of the new environment changing the residents' behavior, the residents had destroyed the new environment. One resident commented: "It was a luxury apartment, but it ended up being a penitentiary!"
What lesson can be learned from the rise and fall of Pruitt-Igoe? In hindsight, observers concluded that the main problem, in addition to architectural blunders, was the lack of participation afforded residents in designing and governing their "city." Accordingly, the St. Louis administration now allows the residents of two new projects, Cart Square and Darst, self-government in running these projects. Modest salaries are paid to certain tenants to serve as liaisons between tenants and project management. An elected board of directors hires all staff, from the overall manager to custodians. Improvements in the buildings can now be made, as rent collections average over 100 percent, with many tenants paying off back rent. Through participatory management, many problems are being solved. The Darst project does lack residents in' about one-third of its apartments owing to a lack of funds to make them livable again; thirty million dollars are needed for repairs. But the experience of the Darst project suggests that the debacle of Pruitt-Igoe was unnecessary. And it suggests that participatory management can be a vital force in organizational effectiveness.
The Pruitt-Igoe project illustrates many other principles of organizational communication. By the time you are finished with this book, you will hopefully be able to recognize these other communication concepts. We aim to provide you with some of the tools for analyzing communication problems in various types of organizations.
The Importance of Organizational Communication
Most of one's daily life is spent in organizations: schools, businesses, factories, hospitals, the military services, churches, social clubs. We live in an organizational society. Imagine that you had the task of collecting a dime each week from your four best friends. Easy, you say. Quite right -- no organization is needed to accomplish the task. But now suppose your responsibility is to collect income taxes from the millions of individuals in your nation. Obviously, an organization would be necessary (Blau and Meyer 1971, p. 3).
Almost everyone belongs to one or more organizations. And most would agree that it is communication that gives life to an organizational structure. One of the early scholars of organizational behavior, Chester Barnard, recognized that "in any exhaustive theory of organization, communication would occupy a central place, because the structure, extensiveness, and scope of organizations are almost entirely determined by communication technique" (Barnard 1938, p. 8). More recently, Katz and Kahn (1966, p. 224) noted that "communication is...a social process of broadest relevance in the functioning of any group, organization, or society." It is "the very essence of a social system or an organization" (Katz and Kahn 1966, p. 223). Herbert Simon (1956, p. 109) probably stated it most sweepingly: "The question to be asked of any administrative process is: How does it influence the decisions of the individual? Without communication, the answer must always be: It does not influence them at all."
An organization is a stable system of individuals who work together to achieve, through a hierarchy of ranks and division of labor, common goals. The relationships among the members of an organization are relatively stable; this structural stability enables an organization to function effectively in accomplishing certain objectives. Organizational structure lends predictability and stability to human communication, and thus facilitates the accomplishment of administrative tasks.
How different is human behavior in organizations from that occurring in other, less structured situations? "The behavior of people in organizations is still the behavior of individuals, but it has a different set of determinants than behavior outside organizational roles" (Katz and Kahn 1966, p. 391). Most of these distinctive determinants are variables that involve organizational structure. For instance, assume that individual A is trying to persuade individual B to use a new reporting form that B does not feel is necessary. Ordinarily, we would not expect B's behavior to change. But what if A were company president and B a lathe operator? Structural variables like hierarchical status and authority make a difference!
A critical reason for studying organizational communication is that it occurs in a highly structured context. An organization's structure tends to affect the communication process- thus, communication from a subordinate to a superior is very different from communication between equals.
Communication is the lifeblood of an organization: if we could somehow remove communication flows from an organization, we would not have an organization. Communication pervades all activities in an organization, represents an important work tool through which individuals understand their organizational role, and integrates organizational subunits. From an open system perspective, an organization is an elaborate set of interconnected communication channels designed to import, sort, and analyze information from the environment and export processed messages back to the environment. Communication provides a means for making and executing decisions, obtaining feedback, and correcting organizational objectives and procedures as the situation demands.
One usually thinks of industrial production, for example, as compromising the application of heat, chemicals, and other materials to an input in order to create an output, but communication is seldom counted as a crucial element. The physical transformation is important, of course, but communication is also very much involved. Communication is a thread that holds the various interdependent parts of an organization together. The functions of planning, coordination, and control are very important processes. If communication were somehow removed from industry, it would collapse instantly. "When communication stops, organized activity ceases to exist. Individual uncoordinated activity returns" (Hicks 1967, p.
Not only is communication an essential ingredient in the internal functioning of an organization, but, as we noted, it is also vital in the organization's information exchanges with its environment. "The communication system serves as the vehicle by which organizations are embedded in their environments" (Guetzkow 1965, p. 534; emphasis added).
The tools of communication research are useful in seeking to understand organizational behavior. "If we can map the pathways by which communication is communicated between different parts of an organization and by which it is applied to the behavior of the organization in relation to the outside world, we will have gone far toward understanding that organization" (Deutsch 1952). We proceed, then, under the assumption that communication is an indispensable element in an organization's functioning.
THE HOSPITAL AS AN INFORMATION SYSTEM
A perceptive visitor to a hospital would observe that a great deal of staff activity is devoted to information processing activities.
At the nursing station are several nurses filling up patient record forms, talking on the telephone, or discussing the coordination of their day's schedule. A doctor is being paged on the P.A. system. Food trays are being wheeled down the corridor to patients' rooms, each tray supposedly assigned to a particular patient on the basis of particular dietary needs, as indicated on a form that was sent previously to the hospital kitchen. At bedside, a doctor is obtaining information from a patient about symptoms, information that will then be entered in the patient's record file.
It would be hard to overestimate the importance of this file to the hospital. If the record file were ever lost or misplaced, as sometimes happens, the hospital would be at an almost complete loss as to how to treat the patient. The file is begun when the patient enters the hospital's door, and it must accompany the patient wherever he or she goes: to X-ray or to surgery, for example. Upon discharge, the file, by this time an inch or two thick with forms, charts, and notes (and representing hundreds of hours of skilled expertise worth thousands of dollars), goes permanently to the hospital records department. The record file is so sacred to the hospital that the patient is not allowed to see it, except in very unusual cases, and then only with the permission of the doctor. The file can be so thick that it represents an information overload problem. One solution is to computerize the file data; another approach is to utilize a "problem-oriented" record system, in which essential data are isolated and instantly accessible.
Great effort is expended by the hospital to ensure that the patient record file contains adequate and accurate information. Each entry in the file must be signed and initialed, so that responsibility for it is precisely fixed. Every bit of information about the patient must be recorded: every, bowel movement, every medication that is administered, every rise in temperature. The cost of an error in the patient file is awesome: each year one or more patients' lives in a typical hospital may be lost owing to errors in information processing or transmittal. Because of the cost of such errors, most hospitals expend great effort on continually redesigning their communication system, on training and retraining the hospital staff to use it properly, and on checking on its accuracy and adequacy through various feedback devices.
Despite such extensive precautions, however, embarrassing communication breakdowns still occur. One spectacular example is the case of a patient in the Chicago-area Veterans Administration hospital, Erwin Pawelski, who was "lost" for twenty-seven hours.
Pawelski, a patient who could not speak, was strapped onto a wheel chair by an attendant and wheeled out of his ward to receive occupational therapy at 9:30 A.M. on May 1, 1975. What happened next is anybody's guess. The patient's record file shows a twenty-seven-hour blank. A hospital spokesman later told newspaper reporter,, "There's a presumption that he arrived in the basement for therapy. But we are not positive."
At 7:00 A.M. on May 2nd, Pawelski's wife was called by the hospital to ask if she had removed him from the hospital. When she rushed to the hospital, she found that another patient had moved into his bed. Pawelski's effects had been shoved into a closet.
Pawelski was found by a therapy supervisor who stepped into an elevator in the hospital basement at 1:10 p.M. on May 2nd. The hospital has 3,000 employees, 1,295 patients, and about 700 visitors daily. Pawelski was in one of the main banks of elevators, ridden each day by hundreds of doctors, nurses, attendants, patients, and visitors. "It's unbelievable that there wouldn't be one person during those twenty-seven hours offering to help this man slumped over in a wheelchair. It's a mystery, what happened,'' stated the hospital spokesman.
Twenty days after Pawelski was "misplaced" for twenty-seven hours, he died from cerebral hemorrhage after undergoing brain surgery. The hospital claimed that his death was not connected with any ill effects sustained from the incident.
A hospital can best be viewed, then, as an organization that devotes much of its activity to processing information. So, in fact, do most other types of organizations.
Studying Communication in Organizations
This book is about communication in organizations. What is communication? We define communication as the process by which an idea is transferred from a source to a receiver with the intention of changing his or her behavior. Such behavior may encompass a change in knowledge or attitude as well as in overt behavior. An organization was previously defined as a stable system of individuals who work together to achieve, through a hierarchy of ranks and a division of labor, common goals. Thus, when an organization executive issues an order to a subordinate, he expects it to be obeyed; the purpose may be carried out, or it may not be. But our point here is that the communication was made with the intention of achieving a certain result.
This book is concerned with communication within organizations and that between an organization and its environment -- which together make up what we call organizational communication. Most of the present chapter is devoted to elucidating the concepts we use to describe the process of human communication. Most of what we say in this chapter applies to all types of communication, not only that in organizations. When the communication process occurs in organizations, however, it is highly structured and geared toward formalization and predictability. In succeeding chapters, we shall detail what is distinctive about communication when it occurs in organizations.
MAIN ELEMENTS IN THE COMMUNICATION PROCESS.
The four main components in the communication model are the source, the message, the channel, and the receiver. Because these elements are almost always present in every communication act, this simple conception of communication is often referred to as the "S-M-C-R" model (Berlo 1960). It is a much oversimplified conception of communication, but it will be useful to us in our discussion here as a starting point. In the presentation that follows, we define each of these four elements (S, M, C, R) plus two others, effects and feedback (Figure l-l).
1. Source: The source is the originator of the message. It may be an individual or several individuals working together, such as a television news team. A source may also be an institution or an organization, although even then individuals are ultimately the sources, even though they are acting in an organizational role. The main responsibility for preparing the messages lies with the source.
2. Message: The message is the stimulus that the source transmits to the receiver. It's what the act of communication is all about; it's the idea that is communicated.
Messages are composed of symbols having (for the source and the receiver) a certain meaning. Encoding is the translation by the source of an already conceived idea into a message appropriate for transmission. To encode is thus to change a meaning into a symbol. Decoding is the translation of received stimuli into an interpreted meaning. Receivers thus decode messages by changing the symbol into a meaning. In order to give meaning to stimuli, individuals classify phenomena in categories and give them labels (called "codes").
Many messages are expressed in the form of language symbols, but the symbols may also be nonverbal, such as hand or facial gestures, other body movements, or pictures. Language distinguishes man from other creatures; dolphins and chimpanzees, for example, may have rough communication systems, but they have no true language.
Meanings are references such as ideas, images, and thoughts that are expressed in symbols (that is, in language). For communication to occur at all, the source and the receiver must have at least some minimum degree of prior common experience, some level of shared meanings. At the other extreme, no two individuals have exactly the same experiences; hence the language used (the message symbols) has somewhat different meanings for the receiver and the source. Furthermore, an individual's experience is continuous, so that the meaning of the same message symbols will change over time. Many failures to communicate are due to mistaken assumptions by source or receiver about the meaning of a symbol they have exchanged.
Meanings are relative and open to subjective interpretation. This fact led Berlo (1960) to state: "Meanings are in people, not in the message." He meant that words have no meanings in themselves; their meanings are assigned by the source and the receiver.
Messages contain information, defined as a change in the probability that some alternative will occur in a given situation (Miller 1955, 1965, and 1972). Conveying information thus reduces the receiver's uncertainty about some phenomenon. An example of information is a market research report to an executive that shows that consumer demand for a product will drop 20 percent in the next three months.
Some messages are new to the receiver, and hence represent a stimulus of a kind different from that contained in ordinary messages. An innovation is an idea, practice, or object perceived as new by the receiver. When the message is an innovation to the receiver, such an act of communication is called diffusion, the process by which innovations are communicated to the members of a social system over time (Chapter 6).
3. Channel: A channel is the means by which a message travels from a source to a receiver. It is the path through which the message is physically transmitted. Channels may be classified into mass media or interpersonal channels.
Mass media channels are those means of transmitting messages, such as newspapers, magazines, films, radio and television, that enable a source to reach many receivers. Interpersonal channels are those that involve a face-to-face exchange between a source and a receiver.
One of the most important differences between mass media and interpersonal communication is that feedback is facilitated in the latter. As we indicated in our diagram of a communication event (Figure 1-1), communication is not a one-way flow of a message from a source to a receiver. The receiver also generates information and messages for the source, and, in fact, such interaction is necessary for communication to thrive. We shall have more to say, later, about feedback.
4. Receivers: The most important single element in the communication process is the receiver. Communicators (sources) often forget him. Some sources are source-oriented; an example is textbook authors who write for their colleagues, and go over the heads of their student readers. Some are message-oriented: they know a great deal about their topic, but they do not express (encode) it meaningfully in terms their receivers can understand. Or sources may be channel-oriented, depending so entirely on a particular means of communication that the receiver is ignored. An example is the official in an organization who communicates solely by the distribution of written memos to his receivers; he never uses a staff meeting, even when it would be more effective, or a combination of a written memo plus a staff meeting.
5. Effects: Communication effects are the changes in receiver behavior that occur as a result of the transmission of a message. Hence when we speak of "effective communication," we mean communication that results in those changes in receiver behavior that were intended by the source. There are three main types of communication effects:
1. Changes in receivers' knowledge.
2. Changes in receivers' attitudes, defined as the relatively enduring organization of an individual's beliefs about an object that predisposes his actions. That is, an attitude often (though not always) predicts the action that an individual may take.
3. Changes in receivers' overt behavior, such as voting, purchasing of products, or coming to work on time.
These three changes usually, but not always, occur in sequence; that is, a change in knowledge usually precedes a change in attitude, which precedes a change in overt behavior.
The purpose of most communication, as we have said previously, is to bring about certain intended effects in receivers. Most communication research by social scientists seeks to study the effects of communication, with the aim of making possible more effective communication. A kind of "component approach" is used, in which one (or more) of the elements in the communication process is manipulated to induce greater effects. We shall have more to say later about this predominant approach in communication studies, and much of it will be critical.
Feedback is a response by the receiver to the source's message. The source may take account of feedback in modifying subsequent messages; thus feedback makes communication a dynamic, two-way process.
Feedback may be thought of as messages to the source conveying knowledge of the effectiveness of a previous communication. Positive feedback informs the source that the intended effect of a message was achieved; negative feedback informs the source that the intended effect of a message was not achieved. As such, negative feedback is disruptive of the source-receiver relationship, and it can generate hostility between source and receiver.
From the viewpoint of achieving effective communication, negative feedback is more important than positive feedback. Yet it is more likely to be unrewarded or even punished, because of its disruptive nature. Ever try to tell the boss that his pet idea is not working? More on this in Chapter 6.
In general, the more feedback-oriented a communication process, the more effective it is. Attention to feedback implies an orientation to the receiver, a concern with whether he or she is "getting the message." Such an orientation requires for its implementation some feed forward, information about the receivers that is gained by the source prior to initiating communication with them, and that is used to design communication messages for maximum effectiveness. It is important to know one's audience. If a source has false assumptions about his receivers, his efforts to communicate are likely to be less effective.
False assumptions are more likely to be made when the receivers and the source are dissimilar. If the two are highly similar, communication is easier and more effective, for the source only has to know himself in order to know his receiver. Homophily is the degree to which a source -- receiver pair are similar in certain attributes, such as beliefs, education, or social status; heterophily is the degree to which a source -- receiver pair are different in certain attributes. Communication within a heterophilous source -- receiver pair is generally less effective than that within a homophilous pair; when source and receiver share meanings, communication is more facile. Heterophilous communication, on the other hand, often leads to message distortion, delayed transmission (because of longer reaction time), restricted channels, and cognitive dissonance (when the receiver is exposed to messages that are inconsistent with his existing beliefs). Yet, while heterophilous communication is more difficult, it also has the potential of conveying new information to the receiver, and hence instigating behavior change. We shall have more to say about heterophily in Chapter 5.
We apply the word noise to disturbances in the communication process that interfere with the intended effects of the communication. An example is trying to talk with someone while a radio is playing loud music. One way to reduce noise is to increase redundancy, the repetition of the message or of some part of it. Redundancy is a kind of "noise insurance" in that, while it does not remove the noise, it reduces the disturbing influence of the noise on communication effectiveness. At least half of typical conversation may be redundant.
THE NATURE OF COMMUNICATION RESEARCH
We seek to utilize a communication point of view in this book. Such an approach implies that human behavior change is not capricious or random, if it can be understood from the receiver's viewpoint. Through social scientific methods we hope to be able to gain this viewpoint, to see the communication event through the receiver's eyes (rather than just from the source's viewpoint). Our scientific approach, however, does not have as its sole purpose accumulating an abstract fund of knowledge about human communication; usually, our objective is also to facilitate the solution of important social problems. So communication research is a highly applied type of social science.
Modern scientific approaches to defining communication can be traced back to the 1940s. Prior to that time, communication as an academic field was mainly a matter of teaching the skills of message production. Human communication was considered an art, a practice, The fields of speech, journalism, education, advertising, marketing, literature, and broadcasting, among others, exemplified this approach. But some leaders in these fields saw that mastering the skills of message production, as important as that was, could not guarantee effective communication. Prior to the development of communication as a science, the effects of messages, however well designed or elegant they might be, could seldom be predicted for a given audience.
The application of social science methods and concepts to human communication promised to help improve our understanding of the likely effects of a communication message to a given audience. Among the pioneers in the new field of communication science were the sociologist Paul Lazarsfeld, the political scientist Harold Lasswell, and the social psychologists Kurt Lewin and Carl Hovland. Their work, and that of their students, began to provide a basis for an empirically based and theoretically integrated understanding of communication.
But communication science is a relatively recent intellectual enterprise, and it still has a long way to go. Presently, there are several hundred university departments of communication in the United States and abroad, of which perhaps fifty award the highest academic degree, the Ph.D. The scientific approach to human communication is also taught within most of the behavioral sciences and their applied fields.
So the communication approach to human behavior change seeks to understand the effects of communication from the receiver's viewpoint, as well as the source's, through the application of social science methods and concepts.
THE COMPONENT APPROACH IN PAST COMMUNICATION RESEARCH
Most communication research in the past was oriented to studying the effects of communication on the receivers. A communication scholar using this approach would alter one of the components in the S-M-C-R communication model (Figure 1-1) in order to determine its effects. Thus, a mass medium might be contrasted with an interpersonal channel to assess which is more effective in bringing about behavioral effects among a particular audience of receivers. For example if the message conveyed is an order, it is more likely to be carried out if it is communicated in written rather than oral form. One large American automobile manufacturer instructs its supervisors to give no verbal orders and each is provided with a memo pad on which is stamped "NVO." Employees are instructed not to accept an order unless it is in written form. Why? Verbal messages are impermanent and less accountable, although they may be more appropriate in a crisis.
The communication researcher may experimentally alter some source variable such as the credibility of the source as perceived by the receiver. An identical message about future employment trends, for example, may be attributed with one set of receivers to the company president, and, with another (but similar) audience of receivers to the president of a competitor; the two messages are printed as articles in two different forms of the company's house organ. Communication theory would lead us to predict that the message attributed to the higher-credibility source (that is, presumably, the company president) will have greater effects in changing attitudes toward the issue.
In yet other communication researches, message variables may be altered to test the different effects on receivers. For example, a similar message may be presented so as to involve fear appeals in the case of one group of receivers, but not another. Thus, employees may be told that unless their output increases the company will enter bankruptcy and they will lose their jobs. Another audience of employees might be told of a new bonus for higher production.
In all of these illustrations, one of the variables S, M, C, or R is experimentally altered in order to determine the differential effectiveness of these variables on the receivers. Effects-oriented research of this kind treats communication essentially as something that one person does to another. Such investigations use an audience of receivers as respondents, and their behavior is studied largely for the benefit of the sources.
Such an approach may present serious ethical problems: for example, why shouldn't the researchers study how receivers can immunize themselves from the source's message, so as to prevent its intended effect? Worse, the overwhelming focus on effects in communication research is often antithetical to our theoretical conception of the process of communication. It implies a linear, left-to-right, one-way aspect to the communication event, which is incompatible with our conception of communication as a two-way, reciprocal-exchange process. Despite these serious limitations, the component approach has been a predominant one in past communication research. But in recent years, an intellectual revolt against this dominant approach has begun.
FROM LINEAR MODELS TO A SYSTEMS APPROACH
Communication is a process -- that is, a continuous sequence of actions through time. It is not meaningful to talk about a beginning or an end of communication, because, like all other processes, communication flows like a stream through time. Someone has suggested that all processes should always begin and end with the word "and."
Our paradigm (Figure 1-1) of the communication process is a linear, left-to-right model. As such, it is a great oversimplication of reality. Conveying the notion of communication as process is helped somewhat when we add feedback to the model. Now our model better reflects a concern with events occurring over time, and hence begins to imply process.
Communication is a multivariable, dynamic interplay of numerous elements. Its complexity can hardly be expressed in a linear model with the channel carrying messages from the source to the receiver, like a bucket carries water (Diaz-Bordenave 1972). Such a transmission approach to communication implies taking a message from one place and reproducing it in another, as a telephone company does.
The black-and-white oversimplification involved in communication models like S-M-C-R led to growing interest in conceptualizing communication as a system. This theoretical view put the multivariable complexity back in communication models, and emphasized the synergistic interdependence of the elements in the communication process. The theoretical shift from linear models to a systems approach happened in the late 1960s, at the same time that most social science disciplines were adopting a systems view of human behavior (see Chapter 2).
The linear models imply a mechanistic concept of the communication act, which facilitates understanding because of its simplicity but profoundly distorts reality. Worse, they may imply an autocratic, one-sided vision of human relationships. "[The linear model] assumes an active source operating on a passive receiver via the persuasive monologue. It thus suggests a vertical relationship in which the source will tend to direct or dominate the behavior of the receiver" (Beltrán 1972). Communication according to this model, would therefore amount to a special type of receiver manipulation by a source.
On the other hand, the systems model of human communication assumes a greater degree of equality between the participants in the communication event. Communication is conceived of as a dialogue, in which source manipulation of the receiver may be counterbalanced by receiver influence on the source. The receiver, in any event, is largely free to determine the meaning of the source's message for himself.
The purpose of communication is to commune with, rather than just to persuade or command. Significantly, the words "common," "commune,'' and "communication" have the same etymological root. Communication is the sharing of information. And sharing implies that two or more people do something together, not that one individual does something to another. In this sense, rather than saying that an individual "communicates," it is more meaningful to say that he engages in communication and becomes part of a communication system. So communication is not simply a matter of action and reaction; it is a transactional exchange between two or more individuals (Watzlawick and others, 1967). Communication occurs as a series of exchanges of messages, with each subsequent message building on the previous one. The roles of source and receiver are reversed with each sequential exchange. Thus, once we bring in the process and systems aspects of communication, the linear and unidirectional S-M-C-R model no longer is adequate to describe the multidirectional, transactional nature of the communication act. Then it is appropriate for communication research to ask not, "What does communication do to individuals?" but rather, "What do people do with communication?"
At many points in this book we are forced to follow an implicitly linear and unidirectional model of communication in order to facilitate understanding. But we wish to stress that this one-way, "timeless" conception of human communication is adopted by us for heuristic purposes only. Communication is in reality a synergistic process in which the elements operate in an interdependent and intricate fashion.
BIASES IN PAST COMMUNICATION RESEARCH
There have been, in our view, three important conceptual/methodological biases in past communication research:
1. the lack of a process orientation
2. associated with this, an ignoring of mutual causality among the elements in the communication process
3. a psychological orientation, leading to the shortchanging of structure in the communication process
The Lack of a Process Orientation
Every textbook definition of communication either states or implies that it is a process. Thus one might expect an overwhelming emphasis in research and theory on the conceptualization of communication as process. A recent content analysis of communication research, however, shows that existing research designs and measurements almost never allow analysis of communication as it occurs over time, which would be necessary to adequately explore its nature as process (Arundale 1971). Very few communication researches include data gathered at more than one point in time, and almost none at more than two such points. Hence almost all communication research is unable to trace the change in a variable over time; it deals only with the "present tense" of behavior. Communication thus becomes, in prevailing communication research, an artificially halted "snapshot" rather than a continuous "film" or moving picture.
Why has communication research not dealt more adequately with the change-over-time aspects of the communication process?
1. We lack concepts and propositions which truly reflect a process orientation.
2. Time-series data are expensive to gather, unless one depends on respondent recall, a methodological procedure that is often less than satisfactory.
3. Repeated data gathering over time leads to problems of respondent sensitization (unless one uses unobtrusive and nonreactive measurement methods), as the research itself is a communication process between respondent and researcher.
4. Communication researchers are often pressured by research sponsors, limited funds, doctoral dissertation requirements, and other factors to produce immediate results; this strongly discourages research over time.
So, unfortunately, we define communication as a process, but then proceed to treat it, in communication research, as a one-shot affair. Most communication research designs allow for only cross-sectional data analysis; such designs cannot tell us very, much about the process of communication over time.
Future research ought to emphasize such improved methods as field experiments and panel studies (repeated surveys of the same audience), which, by their research designs, are better able to take "moving pictures" of the communication process as it occurs over time.
The Bias Against Mutual Causality
Previously, we described the component approach in past communication research, in which a source, message, channel, or receiver variable is experimentally varied in order to test the consequence of this manipulation for the effects of communication. Changes in the source, message, channel, or receiver thus constitute the "independent" variables in such research; the "dependent" variable consists of communication effects which are caused by the change in the independent variable.
In order for independent variable X to be the cause of dependent variable Y: (1) X must precede Y in time; (2) they must be associated; and (3) X must have a "forcing quality" on Y. Most past communication researches have determined only that various independent variables are associated or correlated with a dependent variable; correlational analysis of one-shot survey data does not allow the determination of time order. So again we see the importance of research designs that allow us to learn the over-time aspects of communication. Field experiments are ideally suited to the purpose of assessing the effect of various antecedent independent variables on the consequent dependent variable.
In order for X to cause Y, they must at least be associated, or vary together. If such covariance is very low, X is probably not a cause of Y. If their common variance is high, X may be a cause of Y.
Forcing quality, the way in which the cause X acts on the result Y, is a theoretical rather than empirical issue, which rests on the inherent nature of the X and Y variables. Much greater attention needs to be given in communication research to the theoretical reasoning of why certain variables might have a forcing quality on others. And the systems perspective in communication research (elaborated later in Chapters 2, 3, and 5) implies that the search for independent and dependent varables in communication research is often futile. The pertinent variables may be interdependent; each is a cause as well as a consequent of the others. They are in reality a system of variables in mutual interaction.
Implementation of the systems approach in communication research would entail abandoning reliance of the past: on the component approach of varying a source, message, channel, or receiver variable to determine its communication effects. Rather, the communication process would be conceived of as a holistic system in dynamic, interplay.
The Psychological Bias
The psychological bias in communication research stems (1) from its historical roots as an academic field, and (2) from researchers' acceptance of how social problems are defined. Several early communication scholars came from backgrounds in psychology, where at the time an individual-centered perspective was dominant. It was only natural that their models of communication largely ignored social structural variables that affect communication. The transactional and relational nature of human communication tended to be overlooked, at least until fairly recently.
One manifestation of the psychological bias in communication research is the overwhelming focus on the individual as the unit of analysis, while the importance of communication relationships between sources and receivers is largely ignored. This distorted approach is often due to the assumption that if the individual is the unit of response, he must consequently be the unit of analysis (Coleman 1958). Coleman (1958) goes on to note that "the kinds of substantive problems on which such research focuses [tend] to be problems of 'aggregate psychology,' that is, within-individual problems and never problems concerned with relations between people."
The use of survey methods in communication research has "destructured" behavior. "Using random sampling of individuals, the survey is a sociological meat-grinder, tearing the individual from his social context and guaranteeing that nobody in the study interacts with anyone else in it" (Barton 1968). The parallel is to "a biologist putting his experimental animals through a hamburger machine and looking at every hundredth cell through a microscope; anatomy and physiology get lost; structure and function disappear and one is left with cell biology" (Barton 1968).
Only recently has the focus in communication research on the individual as the unit of analysis shifted to the dyad, the clique, the network, or the system of individuals; that is, to the communication relationships between individuals rather than the individual themselves. Examples of new approaches seeking to overcome the psychological bias in communication research are the coorientation model, relational analysis, network analysis, and the systems approach (all of which are detailed in Chapter 2 and 5).
These conceptual/methodological approaches assume that even when the individual is the unit of response, the communication' relationship can be the unit of analysis via some type of sociometric measurement (Chapter 5). Sampling and data-analysis procedures for relational analysis are being worked out, but we still lack relational concepts, and theories linking these concepts. Until communication scholars begin to think in relational terms, there will not be much relational analysis.
A second manifestation of the psychological bias in communication research is the acceptance of a person-blame definition or causal attribution of social problems. Person blame is the tendency to hold an individual responsible for his or her problems. Clearly, what is done about a social problem, including research, depends on how it is defined. Seldom do communication scientists participate in the identification and definition of social problems; so they borrow or accept these definitions from alarmists, government officials, or other scientists.
Many illustrations of person blame can be cited in behavioral research. Caplan and Nelson (1973) find a high degree of person blame in psychological research on such social problems as highway safety and race relations. Dervin (1971, p. xiii) noted that person-blame assumptions affected communication research on the problem of poverty:
Only very recently have some students of the poor come to see that it is the social structure, not the poor as individuals, that needs change. It is incomplete, for instance, to say that the poor lack knowledge where the system does not make information available to them.
Person blame rather than system blame permeates most definitions of social problems. The definers are seldom able to change the system (or so they think), so they accept it. Such acceptance encourages a distorted focus on psychological variables in communication research. Often, the problem definer's individual-level cause becomes the researcher's main independent variable.
How can the person-blame bias be overcome? By keeping an open mind about the causes of a social problem, at least until exploratory data are gathered. By involving the participants in a social problem, rather than just those persons who are seeking its amelioration, in the definition of the problem. And especially, in research on communication in organizations, by considering structural variables, as well as intraindividual variables.
Using a systems point of view in research on communication in organizational structures is a step toward coping with each of the three biases in past communication research. First of all, a systems viewpoint (elucidated in Chapter 2) assumes a concern with process; it orients the researcher toward gathering data at more than one point in time so as to take moving pictures instead of snapshots of human behavior. Furthermore, from a systems viewpoint, there is no simple cause-and-effect; all the variables indexing a system are thought to be in mutual interaction. Finally, the systems approach considers the effects of organizational structure on communication behavior, and vice versa (Chapter 4).
RESEARCH ON COMMUNICATION IN ORGANIZATIONS
Much attention in the literature about organizations has been given to communication problems; in fact, reading these publications gives one the impression that insufficient communication is frequently the main cause of difficulties in organizations. We dispute this overly simplified picture of organizational communication problems (in Chapter 4), and, furthermore, we doubt the intellectual usefulness of the past focus on communication problems in organizations. More important is the ability to understand the factors causing such problems. We agree with Thayer (1967, p. 80), who observed: "Perhaps more has been 'communicated' about 'communication problems' in organizations than [about] any other single topic in the field. Yet this plethora of commentary has not been conducive either to theory'-building or to theory-validation.."
Most early communication research in organizations consisted of case studies of a single organization. Researchers immersed themselves for a considerable period of time in one organization, observing what went on in its everyday activities and perhaps administering questionnaires to the organization's members or personally interviewing them. The researchers might focus on one department within an organization or even on a small work group within a department. This intensive approach enabled the researchers to learn a lot about very little. But it did not provide a very sound basis for generalization of the research results. For example, how was one to know whether the communication behavior of the fourteen workers in, say, the Bank Wiring Room of the Western Electric Hawthorne plant was at all representative of the thousands of employees in the Hawthorne plant, let alone of all assembly-line workers in all factories in the United States?
Organization scholars subsequently began to conduct comparative studies in two or more organizations, In order to handle the larger amounts of data now involved, researchers were forced to learn less about more, and to convert their impressions into statistical data that could be punched onto IBM cards and analyzed with computers.
Finally, communication researchers began to gather data from larger samples of perhaps a hundred or more organizations. This approach provided a more solid basis for generalization of the research findings, but it usually necessitated knowing very little about any one organization in the sample that was studied. Frequently, a questionnaire was sent to the chief executive of each organization sampled, so that only the view from the top was obtained. This extensive, quantitative approach essentiallydestructured the organization, in that relatively little about the internal nature of each organization could be measured.
One problem in generalizing the results of scientific research in organizations was that in the past, such studies were mainly concerned with a rather narrow range of industrial, business, and governmental organizations. Modern organizations have certain similarities whether they are concerned with the production of a product for profit, the conduct of governmental business, a charitable enterprise, the education of students, the treatment of the ill, or the defense of a country. Now attention is also being given in organizational communication research to such other kinds of organizations as schools, hospitals, prisons, labor unions, cities, voluntary organizations, and political parties. Nevertheless and regrettably, we know a great deal more today about organizational behavior in business firms and governmental bureaucracies than we do about other kinds of organizations. "The literature leaves one with the impression that after all not a great deal has been said about organizations, but it has been said over and over in a variety of languages" (March and Simon 1958, p. 5).
"Organization theory...has been altogether too accommodating to organizations and their power" (Perrow 1972, p. iii). Most research in this field has undoubtedly benefited management more than it has lower-level workers, and members of the organization more than its clients. There are many reasons for the pro-organization orientation in research on organizational behavior, including who sponsors the research, for whose benefit it is conducted, and what methods of study are used. In this book we seek to evaluate research on organizational communication with an objectivity that will hopefully make our findings equally helpful to the executives, the workers, and the clients of an organization.
TOWARD NETWORK ANALYSIS
The general trend in research on organizational communication behavior has been toward greater quantification, toward reducing human behavior to variables that can assume numerical values. This use of highly quantitative measures in social science research may be deceitfully reassuring to the researcher. This may be one reason why questionnaires and survey interviews have been so widely utilized in studies of organizational behavior. Writing in 1934, a leading social psychologist warned that "the questionnaire is cheap, easy, and mechanical....Quantitative measurements are quantitatively accurate....Yet it would seem far more worthwhile to make a shrewd guess regarding that which is essential, than to accurately measure that which is likely to prove quite irrelevant" (LaPiere 1967, p. 31).
Unfortunately, survey research methods are not entirely satisfactory for studying communication in organizations. One reason is that interviews or questionnaires tend to isolate the respondent as an atomistic entry, at least in a heuristic research sense, while the very nature of organizational communication is relational, holistic, and structured. Survey data can, of course, be gathered from individuals in an organization, but how does one take into account the position of the individual respondents in the organizational structure? In a random sample of 100 respondents from among the 1,000 members of an organization, what if the organization's president is randomly selected? Should his responses be weighted equally with the janitor's? And how are such relational phenomena as communication flows and patterns measured when the data represent only individual responses?
One important modification in survey research procedures is to ask sociometric questions about communication behavior. An example would be: "With which other individuals in this organization have you talked within the last day?" The responses lend themselves to such types of relational analysis as communication network analysis. Thus, even though the individual remains the unit of response, he need not always be the unit of analysis.
One approach to research on organizational communication has been to discover and analyze the interpersonal communication patterns within an organization (or within some section of an organization) with network analysis procedures, and then to compare these patterns with the formal communication patterns that would be expected on the basis of the organizational structure. Network analysis is an essential research tool in analyzing the existing reality; specifically it is used to identify cliques within the organization and to isolate such key communication roles as liaisons and opinion leaders (Chapter 5). In the case of bürolandschaft or "landscaped offices," network analysis is utilized to guide the physical arrangement of individuals and hence indirectly' to reorganize the formal structure of the organization so that it is more closely in line with the reality of the informal communication structure.
In any event, comparison of the informal communication patterns with the formal structure illuminates the discrepancies between the two, and this procedure can be a useful diagnostic tool, leading to suggesting changes (1) in the formal structure, to bring it more into line with the interpersonal communication patterns, or (2) in the physical layout of the organization, so as to indirectly change the informal and/or formal communication patterns.
We return in Chapter 5 to a more detailed discussion of how network analysis procedures can be used to understand and improve communication behavior within organizational structures.