Copyright 1997, JD Eveland. All rights reserved.
GLUE, LUBE, AND MONEY: ALTERNATIVE METAPHORS FOR MAKING SENSE OF ORGANIZATIONAL INFORMATION AND COMMUNICATION
Working Paper, California School of Professional Psychology
The process of making sense of all the manifest oddness of how organizations communicate and use information is singularly complicated by the fact that such descriptions must be conveyed in the same medium that is responsible for introducing much of the complexity -- that is, ordinary language. Such language is a frequently blinding combination of literal reference and figurative expression -- the latter to the degree that, as Lakoff and Johnson (1980) expressed it, we "live by" it. Figures of speech are so embedded in the framework of our terminology that we are often unaware that we are using them -- or the degree to which our ideas, reactions, and formative utterances are shaped by the association of such figures.
Analysts of organizational communication and information processing are as affected by this phenomenon as anyone else. When we try to dissect the structure, processes, and function of information flows and effects, we must employ the same kinds of figurative language that we use to describe our most ordinary experiences (Marshak, 1993). Despite the urging of some analysts (e.g., Pinder and Bourgeois, 1982), it seems fairly clear that there is no way to avoid the use of linguistic figures -- and it would be a bad choice to try. The real choice is simply how well and how consciously we employ them (Morgan, 1982). Thus, we continually face the challenge of finding ways to express our understanding that will suggest as well as denote what we intend to convey.
Another complicating factor is the degree to which organizational information processing is in fact the defining feature of organizations (Schall, 1983). That is, there is no way to effectively separate a description of any organizational phenomenon from a description of the ways in which the organization exchanges and manages information. Take away information and communication, and what is left? A conglomerate of individual persons each pursuing personal purposes, a mix of machines, buildings, and bank accounts, a legacy of paper. Coordination, interaction, common direction -- virtually all the things that give meaning to the term "organization" are in fact properties of its communications systems. Thus, we are reduced in large measure to a kind of circular definition, in which organization is communication and communication is organization. This is not to suggest that there are not aspects of either topic that can be expressed in language not directly derivative from the other. It is true, however, it is difficult to make many statements about organization that do not involve at least implicit statements about information and/or information processing.
The concepts of "communication" and "information" are themselves inextricably entangled. It is beyond our scope here to attempt comprehensive definitions of these phenomena -- still less to attempt to distinguish the terms unequivocally. For the moment, let us define information to be the part of human understanding that reduces uncertainty about relationships of phenomena. Communication is the set of processes by which such information is shared between units in a system. When we talk of information, therefore, our emphasis is on the content or meaning associated with the term; when we talk of communication, our emphasis is on the processes of sharing. Neither term makes sense, of course, without the other. Meaning that is not shared is hardly meaningful; sharing without content is equally problematical.
The purpose of this review is to discuss some aspects of the use of language figures as they affect our understanding of communication and information phenomena. If we are more or less compelled to use various figures an any such analysis, it behooves us to approach figure usage with as much care and attention as we would give any other part of the process. In particular, we will be concerned with several possible metaphors that have been used to describe communication systems and information, and with their advantages and disadvantages. We will conclude with some further comments on the degree to which our difficulties are compounded by the fact that all the metaphors we will discuss are true -- while, of course, also being essentially false. This apparent paradox can, we believe, illuminate several aspects of why so many things about organizational communication seem to make little apparent sense.
Metaphoric Language and Reality
As we have noted, metaphor and related speech tropes making up the generic set of analogy are an inseparable part of language use, and it is vastly more difficult -- often impossible, certainly less expressive -- to try to understand or share ideas in language free of metaphor than it is to use them. Technically speaking, there are a wide variety of specific variations on the theme of metaphor. We will use the term in a broad sense, referring to any variety of language in which terms literally descriptive of one object or phenomenon are used to convey meaning about another object or phenomenon to which they do not literally apply. The point is that such non-literal descriptions supply the referents to other phenomena that show off particular elements of the phenomenon in question. To employ an analogy to illustrate the principle of analogy -- it is as though we seek to show off all sides of an object by reflecting it in a series of mirrors such that only slight shifts in glance and focus can give us a view of any given facet. Thus, a metaphorically rich description offers the possibility of demonstrating far more aspects of things than does one that concentrates entirely on the literal.
Erickson (1990) outlines five criteria for determining the effectiveness of a figurative description of a phenomenon: 1) how well structured the metaphor is; 2) how relevant it is; 3) how easy is it to represent; 4) how suitable is it to the intended audience; and 5) how easy is it to extend it to other situations. No single metaphor is necessarily equally effective on all criteria simultaneously. The target is a representation that balances its performance on all criteria -- or, alternatively, is so compelling in terms of one or two criteria that the others no longer matter.
It is beyond our scope here to address the question of whether or not there is in fact anything appropriately categorized as a purely "literal" or "objective" description. The easy postitivist assumption of the existence of "objective reality" has been challenged by a wide range of phenomenological perspectives that postulate various versions of individual or social construction of reality. For our purposes here, let us work with the concept that there is a sort of "consensus literality" that is made up of all the parts of things that would be commented on by most viewers. While this literality may contain some analogical elements, it is capable of being as free of them as the nature of language permits.
To illustrate: a glass of red wine can be defined more or less literally as a red fluid of a certain degree of acidity, containing various esters and dissolved solids, with a certain alcoholic consistency and limited nutritive value. While such a description would probably be agreed to by most observers, it falls remarkably short of providing a satisfying description of the stuff consumed at, say, the monthly tasting by Les Amis du Vin. In that context, one is more likely to hear a description containing referents such as "full of vanilla", "oakey", "with a touch of heather", "big-bodied", "long-lasting", and related ideas that, while they enrich our understanding of the experience of consuming the fluid, are scarcely the common property of all observers.
Neither description is necessarily "true" or "false". Rather, they are of varying degrees of utility depending on the observers and the context. The description that is accessible to and meaningful for the members of Les Amis du Vin is not necessarily the one valuable to say, a non-drinking member of the American Chemical Society. Likewise, various more or less metaphorical descriptions of organizational communication and information convey different kinds of emphases that are appropriate and useful in different situations and audiences. In the next sections, we offer a number of metaphors for information and communication phenomena to illustrate this point.
Before taking up this thread, it is critical that we also understand some of the very real problems and limitations posed by the use of metaphoric language. The first problem is a simple extension of the use of any specialized language, jargon, or other variety of verbal shorthand -- that is, one may use terms that simply have no common referent with one's listeners. While it is efficient for those inside a large organization, say, to employ acronyms for the different components of the system (DoD, USFS, PPRO, J-1, and GPSS are all meaningful in certain times, places, and communities), there is no doubt that this can pose barriers -- desired or undesired -- communicating outside the system. To those outside Les Amis du Vin, referring to a particular Cabernet as "heatherish" is likely to seem irrelevant at best and effectively mystifying at worst. Inside the group, it can be a relatively precise way of specifying a combination of flavors often detectable only by a relatively experienced palate. Thus, the utility of "heatherish" as a description is limited by the ability of the hearers to identify the term with experiences that they can themselves access out of their own repertoires. Neilson's (1992) metaphor of the organizational strategy process as a "jazz band" further illustrates this issue; the "tonally challenged" among us, including your author, who have no idea at all what a jazz band does, tend to miss the point.
A second limitation of metaphoric language is the problem that the same metaphor may convey significantly different associations to hearers with different experiences of the phenomenon referred to. If the eminent organizational psychologist Dr. John-Boy Walton were to describe a certain organization as behaving "just like family", it is likely that this will convey a rather dissimilar set of associations to, say, Dr. Puggsley Addams. One taster's "heatherish" can easily be another's "smoky". If, as below, I refer to organizational communication as "like a web", an arachnophile is likely to associate this image with patterning, skillful construction, and meticulous attention to connectivity; the arachnophobe may well emphasize clinging stickiness and the malignant designer lurking in the middle. Clearly, all these examples are overdrawn exaggerations -- but only by a matter of degree. The problem of varying degree of similarity in associations of a given image cannot be avoided, and the richer and more evocative the association, the greater the probability of significant differences.
A third limitation of metaphor is its inherent narrowness. For each new perspective we provide on a phenomenon through language figures, we also simultaneously deemphasize a variety of other perspectives. To the extent that an organization is "a machine", it cannot also be at the same time "a brain"; the two figures simply give rise to too many contradictory analogic properties. Thus, the use of metaphors in combination is usually more an exercise in creative contrasts than in synthesis.
To summarize, we need to employ all of Crider and Cerillo's (1991) four varying "strategies" for interpreting metaphors. The first, metaphor as "vivid and concise communication", emphasizes the element of in-group communication efficiency to which we have referred. The second, metaphor as "creating or revealing something new", targets the possible use of figurative language to enrich the experience of others. Metaphor as "transforming a perspective" emphasizes our ability to enrich our own interpretations, while metaphor as "simultaneously representing multiple interests", looks ahead to where we will explore the utility of multiple images in combination.
Let us now turn our attention to a variety of metaphors that have been applied to organizational communication and information processing. The listing makes no pretense of being complete, or even representative of the full range of figures that can be or have been used. However, it probably reflects the general types of tropes employed. We begin with three metaphors for organizational information, then go on to describe several classes of metaphors that have been applied to communications systems. We conclude by suggesting that approaching metaphors in combination can shed new light on a number of perplexing aspects of organizational behavior.
Metaphors for Information
The metaphors here treat of communication as glue, communication as lubricant, and communication as money. None are unique to this review; neither do they purport to be a complete catalog of possible figures. However, they do seem particularly useful in their implications.
Glue: It is reasonably commonplace to speak of the pattern of communications as the essential glue that holds the organization together. Several varieties of the "organizations-as-communicatively-constituted" perspective were alluded to earlier. "Glue" is the figure for the pattern of interconnections, interdependencies, and commitments that characterize complex systems and keep them coordinated toward the achievement of complicated ends that require joint action by many parties. There is no doubt that this is much of what organizations do, and thus the metaphor highlights an essential aspect of what "being organized" is all about.
But the implications of this glue metaphor are seldom fully understood. One of the implications is that all parts of the organization are in some sense glued to all other parts. Yet we know that in practice, many organizations operate in an extremely loosely coupled fashion (Weick, 1976). Moreover, there is no inherent benefit in being "glued"; many parts of organizations would probably operate just as effectively, perhaps more so, if they in fact came unglued. Glue ties things down, impedes movement, fastens the pace of the fastest moving parts of the organization to that of the slowest. Actually, this is what happens over and over again -- at least often enough to suggest that the glue metaphor can explain some of the significant negative experiences of organizations as well as the positive ones.
Glue is a simple metaphor, not nearly as rich in its explanatory power and imagery as some others. Yet it does highlight a critical function of organizational communication in a homely and graphic fashion -- one that is instantly accessible and relevant to anyone who has ever behaved organizationally.
Lubricant: Another simple and often employed metaphor suggests that information is a form of lubricating oil -- a substance that surrounds the functioning parts of the system, insulates them from one another and yet enables them to work together without the friction that would otherwise wear them down and grind them to a halt. The lube metaphor is clearly associated with a "machine model" of organizational process (Morgan, 1986), and has its clearest analogic equivalents in that context; however it is not wholly limited to that imagery. The idea of communication as "oil on troubled waters" is a related figure. The presumption is that organizational issues become complicated when messages are not exchanged, and that additional communication is generally a mechanism for reducing that disagreement and miscoordination.
Information as "lube" is again a simple and powerful image that clarifies for us many aspects of why we bother to communicate. Painful thoughts of the cylinders in an engine melting to the walls in the absence of a film of oil are graphic enough to attract and hold our attention, and the organizational analogues -- units creating contradictory strategies, undermining each other, failing to coordinate -- are easy to visualize. The parallel extension -- that even a thin film of oil, a relatively small amount of communication -- can be enough to alleviate the problems is also key.
The image also suggests to us another aspect of communication -- the fact that we are often aware of it only through its absence. The engine takes its oil supply for granted, and only protests when it is somehow not in evidence. Likewise, the organization usually assumes that communication will take place, and reacts strongly only to significant failures in information sharing. As organization members we live constantly bathed in an information environment in which we are less conscious of flows in and out of our space than we are of the individual items that surround us continually. But just as the oil in an engine does not change itself (however much we beg), the information supply available to the organization does require external input.
Like any other figure, the oil metaphor has its disadvantages. Most particularly, its imagery encourages us to focus on the machinelike properties of the organization rather than its more organic, humanistic, and cognitive aspects. In addition, it tends to suggest that all information is somehow equivalent -- and positive in its effects. The more oil -- or communication -- the less the friction, the better the machine works. But it is all to easy to think of instances where communication actually exacerbates tension, where less sharing may make it easier to work together -- speaking metaphorically, where "familiarity breeds contempt". A modest degree of illusion is often helpful in the construction of coordinated goals, in interpersonal as well as organizational relationships. And certainly, unlike quarts of oil, not all items of information are equally important, useful, or relevant.
However, on balance the oil metaphor is graphic and useful -- graphic because it associates images with familiar physical properties, useful because to the degree that organizations are machines, information really does serve the analogous function of lubricant. Used carefully and in conjunction with other metaphors, it can be helpful in getting us to form a more complete view of organizational information.
Money: The third figure in our trilogy of simple images is that of information as money. Again, this is hardly a new insight. Much attention has been devoted to the role of information as a resource. From Williamson's (19XX) transaction-cost analysis to Pfeffer and Salancik's (1978) resource-dependency perspective, the role of information as a commodity to be traded, hoarded, and generally interpreted in market terms has been a commonplace of organizational analysis. This approach has been extremely helpful in exploring both the economics and politics of internal as well as external organizational arrangements.
The metaphor is useful enough to ensure its continued applicability. There is no doubt that many information transactions are explicit or implicit trades. There is no doubt, also, that having information that others do not have usually conveys significant advantage in an environment where competition determines survival. The current debate in American society over what will happen to "information have-nots" in a future that is increasingly information-rich illustrates the power that the metaphor exercises.
Yet for all its utility the money metaphor can be misleading. If its strength lies in its ability to describe the role played by data in an environment where information is a scarce resource, a significant part of its weakness lies in its inability to remain relevant in the same way in an environment characterized by information glut, where the scarce resource is no longer the information itself but the ability to sort, process, and interpret it. The basic mechanics of the management of information scarcity -- control over information channels and flows, what gets into (incoming Metaphor alert!) the pipeline and what comes out, and where -- are significantly less powerful in terms of organizational leverage when the problem is not getting things into the pipeline but keeping them out lest it get clogged.
Again, the metaphor is powerful in describing the organization as an internally competitive system, and there is no shortage of suitable case illustrations. But to the degree that harmony and coordination reign and internal competition is low, the metaphor loses explanatory value. Defining why what is valuable in one context loses that value in another requires explanations rooted in neither economics nor politics, but psychology or even philosophy. Within the money metaphor it is altruism, not self-centeredness, that requires explanation as abnormal . But most organizations are characterized significantly large portions of the time by a subordination of individual and group interests to those of the larger collectivity. If all organizations did with information was trade it, things would come to a rapid halt. In fact, it is largely through such "prosocial organizational behavior" (Brief and Motowidlo, 19XX) that organizations accumulate the resources to function. A transaction-based view of organizational communication is baffled by this behavior.
The concept of information as money subject to internal and external market transactions is thus, like all the other metaphors we have considered, useful but limited. It helps us understand why information often does not flow as less restrictive models would have us to understand that it should. But it also establishes a set of criteria for determining when information flows should take place that simply do not operate in many cases. It is most useful in accounting for organizational failure rather than success; its most positive contribution is a warning that organizations need to establish policies to curb the inherent evil of their members. Analysis of the economics of organization, like that of the economics of societies, is justly seen as a "dismal science".
Metaphors for Communication
In this section we outline some types of metaphors that have been applied to the analysis of organizational communication systems. These metaphors are broader in scope than those applied thus far to information; in fact, each class has several specific variants of somewhat differing emphasis and utility. But collectively they probably express the most useful approaches.
Transportation: One large class of metaphors for organizational communication shares the common property of emphasizing the transportation properties of organizational communication -- that is, a system for moving things from place to place. Various instances of such figures emphasize and suggest somewhat different aspects of the generic metaphor, particularly the content of what is being transported and the nature of the channels involved. They all share most of the same sets of advantages and limitations.
Probably the most overused communication trope today is that of the "information superhighway". In general, "highway" and "road" metaphors suggest that the key role of the communications system is maintaining a set of channels in good repair over which information flows in "packets" from one point in the system to another. This metaphor is thus the direct expression of the "source-message-channel-receiver" model (Shannon and Weaver, 1949) that has served to guide so much exploration of the geography and function of information systems. In addition to the most common version, the "road", transportation metaphors also include variations such as the "conduit" or pipeline, and the "electrical grid". One of the more intriguing variants on the transportation model was suggested some years ago by one of my students -- the "ant farm" dearly beloved from childhood, with industrious little packets of data skittering around in the channels.
There are several more aspects of this metaphor that deserve attention, however. The first is the suggestion of design -- that is, that the system was constructed by human beings for essentially rational purposes. No one builds highways by accident or in a fit of absent-mindedness. They are capital investments, made by the organization to achieve long-range goals. Second, highways require regular maintenance. Without steady inputs of resources and energy, they fall into disrepair and disuse. Moreover, they require usually expensive support systems, sub-organizations devoted entirely to their care and feeding and promotion.
Most significantly, transportation metaphors emphasize the separate identity of the communication "channel", the "message", and the end points ("source" and "receiver"). Road metaphors explicitly suggest information packages analogous to cars and trucks; conduit and electrical metaphors are less parcelized, but still suggest flows of information inside channels that have an entirely separate and prior existence.
Transportation metaphors have obvious advantages, but carry with them some largely undiscussed extra implications. In fact, many information systems, built out of miles of wires, transmission facilities, and receptors do look a lot like roads, and no one will easily confuse the OJ stories passing over the wires with the wires themselves. As noted, much of communication theory takes this metaphor for granted, and such theories have been highly productive in terms of system design, routing optimization, load analysis, and related problems.
The less helpful implications are harder to identify. As Axley (1984) has correctly noted, conduit versions of the metaphor tend to systematically underestimate the costs and difficulties of operating information systems. In addition, the idea of roads usually implies that one can get from any point to any other point, given enough investment of energy. All roads are implicitly connected to all other roads (Tolkein, 19XX); in practice, not all communication systems are connected to all others. In addition, the capital investment aspects of the road building model also imply that building the road and sending things along it are different classes of expenditures, and should be accounted for differently; this may encourage less than optimal allocation of design and operating expenses.
Finally, transportation metaphors emphasize point-to-point transactions, rather than reciprocality. While more modern versions of Shannon and Weaver's theories have added "feedback" loops of various forms to the basic model, they have always been a slightly awkward afterthought rather than an integral part of the approach. The strongest aspect of transportation models of communication is their effective highlighting of purposive messaging behavior. That not all information sharing takes this characteristic is also the models' greatest weakness.
Another significant limitation is that these models tend to be reductionistic is character. That is, they induce us to concentrate our attention on particular roads linking particular points, and the packets of information flowing along them, rather than on the entire pattern of relationships. One reads a highway map to figure out how to get from Los Angeles to Bar Harbor, not to appreciate the intrinsic inner beauty of the entire pattern of the Interstate Highway System. This limitation is to some degree overcome in the next general category of models.
Space: Another general class of organizational information metaphors can be termed "spatial" in nature. These metaphors conceive the organization's information to be configured in the form of a number of compartments or "rooms", with varying degrees of accessibility among them. Different rooms involve different kinds of activities. The flow of information between rooms is partly a function of propinquity and partly a function of necessity. The entire structure is presumed to have some degree of permanence, although the configuration of individual rooms may take place more frequently; as in government offices, remodelling can be a fact of life.
Benford et al. (1993) present an interesting instance of a spatial metaphor. In their framework, computer-based "chat rooms" can be linked into "buildings" and in turn into entire "information cities". Certainly the employment of spatial metaphors for structuring talk in online networks and IRC channels has a good deal of appeal, and the explicitly spatial and even visual structure of "virtual worlds" such as MUDs and MOOs emphasizes the utility of this approach (Turkel, 1995; Stone, 1995).
In fact, the use of spatial metaphors to structure information processing in the computer environment is pervasive. The "desktop" of both Macintosh and Windows interfaces, with its "folders" of data and icons representing physical tools such as calendars, calculators, and trashcans, echoes our surrounding offices and cubicles, while "cutting and pasting" analogizes what we used to do to paper (Wozny, 1989). Most current "groupware" tools employ metaphors such as the "electronic whiteboard" to describe their shared information-processing space (Mark, Haake, and Streitz, 1996). Such structuring reflects the explicit or implicit equation on the part of system designers of computing environments and the offices within which they see them located --- perhaps as an unspoken antidote to the idea of computers as less-than-serious game machines that they often took on before the PC explosion of the later 1980's.
Brain: "Neural" or "brain" metaphors for organizational communication resemble spatial metaphors in a number of ways, although the introduction of the cognitive element introduces another element which warrants separate consideration. Neural metaphors emphasize the similarity of the organization's information processing activities to those of the brain and nervous system. Like the transportation metaphor, the neural metaphor implies purposiveness and direction to the process of communication, although in this case it is a self-generated purpose, not purpose imposed be an externally structured design process. The implication of considering organizational information processing as cognitive in nature is that the organization is somehow more than the sum of its parts, that it takes on some self-directed character as a result of the interplay of its elements, much like Hofstader's (19XX) hypothesized "intelligent anthill". There is a clear connection between the employment of neural metaphors and a gestalt psychology of organizations.
Neural metaphors have great appeal, but also (predictably) pose some problems. Part of their appeal is that they reflect the complexity that we see in real organizations; however complicated the map of the Interstate Highway System (or even the Information Superhighway) may appear to be at a glance, they represent only a fraction of the interconnective nuances of even a modestly-sized organization -- let alone a brain. A neural metaphor can reflect the multiple dimensions of communication, its ability to be many things simultaneously, and the constantly reflexive qiuality of message transactions. Its limitations, aside from the philosophically debatable implication of organizational gestalts, reflect those of the analysis of cognition generally -- particularly, our inability to parse the complexities involved, to make sense of the "black-box" properties of the units involved. Between neuroanatomy and cognitive psychology the gulf still yawns impassibly (ref that thing of mother's); the neural metaphor of communication, while often descriptively satisfying, is singularly short of prescriptive utility.
Nets: After transportation metaphors, the general class of metaphors that looks at organizational communication as a form of net, a pattern of links joining various parts of the system in ways that may or may not reflect purposive design, is probably the most frequently employed. Net metaphors tend to concentrate on the nature of the links themselves rather than on a separate channel-message distinction. In net metaphors, of which there are again several variants including the "spiderweb", the "matrix", and the "fishnet" (Johanssen and Swigert, 1994), the connections themselves can have many different properties.
Network analysis in all its many different forms (Knoke and Kuklinski, 1982; Wasserman and Faust, 1994) tends implicitly to employ net metaphors. The node and the link are the primary units of analysis. The assumption is usually made that common metrics can be calculated for different types of networks, and that it is in fact not even necessary to know much at all about the subject matter and content of networks in order to compare their structural and functional properties. After all, spiderwebs are all different in detail, but largely similar in purpose; one web looks pretty much like all other webs.
This universality of nets is both an analytical strength of the metaphor and the source of some complications. The strength is conveyed by the accessibility of a wide variety of powerful analytical strategies for looking at the overall pattern of webs and examining in detail their wholistic properties. The properties of individual nodes that such metaphors encourage us to focus on are those that reflect the relationship of the node to the entire pattern, such as centrality, integrativeness, and connectivity. Networks are mapped through techniques such as hierarchical cluster analysis and multidimensional scaling, techniques that emphasize the construction of the entire pattern from the relationships of all elements to all other elements. No net model is built up out of isolated links; rather, the links are interpreted collectively to construct the whole (Wasserman and Faust, 1994)
One major advantage of the net metaphor is also one of its greatest weaknesses -- namely, its universality. To the degree that we emphasize the structural/functional similarity of different kinds of communication systems, we tend to underestimate the uniqueness conveyed on each setting by virtue of its content and context. We tend to convey the illusion that, for example, centrality in one setting is mappable in some sense onto centrality in another setting. While this is certainly true at the level of the metaphor, it is almost certainly not true at the level of the real system. Bill Clinton's position relative to the Federal government is only in the most general terms imaginable equivalent to the position of a queen bee in a beehive, although they are both certainly the recipients of great quantities of units buzzing around them. Thus, the metaphor tends to simplify relationships at precisely the point where they probably need to be understood in more complex senses.
The net metaphor in all its various versions, then, is useful -- just like the transportation metaphor. But what both of them tend to lack, in their concentration on the mechanics of the relationships that make up communication behavior and information exchange, is a sense of essential insight -- a pure and stimulating comment on why all the buzzing around, why all the little packets switching back and forth. Metaphors of communication and metaphors of information are often used without a great deal of thought. Below, we offer some suggestions on the effects of treating metaphors simultaneously.
Combining the Metaphors
Evidently, the value of a metaphor is not in its truth or falsity, but rather in its usefulness relative to the behavior we are trying to account for. No single metaphor is powerful enough to account for a great deal. Collectively -- and most particularly in combination -- they can suggest many useful insights. The goal -- as in Lyman's (1995) analysis of computers as "machine, text, and culture" -- is to use one metaphor to illuminate and compensate for another.
Naturally, some metaphors combine more easily than others. Transportation figures of communication require metaphors that will allow for "flow". Either oil or money will do (as will cars, Federal Express packages, or ants) -- glue will not. Net and spatial metaphors, being static rather than dynamic, are compatible with a "glue" view of information (as they are with views of information as electric impulses, subetheric waves, and incorporeal vibrations).
Consider, for example, the value of thinking of organizational information as functioning simultaneously as "lube" and "money". The image that comes to mind is that of a car engine in which the oil pump decides to start charging the cylinders for lubrication -- or the sensors refuse to transmit data on engine temperature back to the pump, in hopes of securing an advantage of some sort. While this image is more funny than compelling, the fact remains that it actually describes many instances of organizational communication breakdown that have serious or even tragic consequences. Information is needed to oil the gears; instead, it is treated by various members or groups as money to be shared or withheld at will, with breakdown the almost certain consequence. On the one hand, this can be castigated as an instance of suboptimization; on the other hand, it can also be seen as an essentially rational (or at least predictably pararational) response to the nature of organizations.
Although we have noted the incompatibility, consider for a moment the complications introduced by thinking of information as "glue" in an environment where the communication system is being analogized as a transportation system. The glue metaphor suggests that information exchange between all parts of the system is important to maximize organizational cohesion and direction. The transport metaphor suggests that individual channels have to be set up and maintained at cost to the system. Whose priorities should govern? Again, the illustration is simple-minded but not unrealistic. Most organizations face implicit trade-offs between the cost of managing communications systems and the benefits derived from them. Despite the urgings of cost-benefit analysts, such tradeoffs are usually resolved in favor of ignoring them, and attempting to maximize both communication and cost control simultaneously, with predictably unpleasant consequences. An uncritical simultaneous embrace of metaphors that suggest diametrically opposite courses of action is as unhelpful as it is common.
Each of the metaphors for information and communications systems that we have outlined in this review is useful and stimulating, and induces us to think about the phenomena of communication in different and helpful ways. The difficulty, as we have suggested here, is that the very fact that all of them are in a measure truly descriptive of what goes on -- while at the same time they often offer rather different behavioral consequences and prescriptions. Many of the peculiarities and seemingly unaccountable instances of organizational communication behavior can be seen as predictable outcomes of the attempt to take two or more metaphors to their logical end simultaneously.
The burden of this review has been to suggest that the metaphorical analysis of organizational communication and information is both necessary and effective -- and also frought with some major problems. Each of the classes of metaphors we have discussed here for both communication and information highlights some critical aspects of the phenomena -- and equally obscures or even contradicts others. Since the use of metaphor and related tropes is unavoidable in organizational analysis, we owe it to ourselves and our clients to use figurative language with care and attention. Effective cooperative problem attention and resolution requires if not perfect, at least compatible, use of language (Krantz, 1990; Burke, 1992)
Our choices of metaphors are not unrelated to the philosophical and epistemological priorities that underlie each of our analyses. Communication metaphors of transportation and brain tend to express positivist values; those of net and space tend to express a more phenomenological orientation. Information metaphors of glue and oil speak to our desire for coordination and harmony; those of money express our reservations that after all there may be a less cooperative dimension to organization.
In the end, the value of any given metaphor is relatively small; the value of metaphorical analysis collectively is significantly great. By picking and choosing our figures of language carefully, we selectively highlight parts of the phenomenon of organizational information, just as (another incoming Metaphor alert!) the analyst at the microscope can approximate the three-dimensional structure of a blue-green alga by turning the focus knob to selectively highlight consecutive planes through the object. No one focal point is truly reflective of the alga; no one metaphor is uniquely valuable in our attempts to untangle organizational communication phenomena. Rather, each one constitutes a valuable tool for analysis -- always remembering the proviso that any tool is a weapon if you hold it right (DiFranco, 19XX). All metaphors are double-edged; use at your own risk!
Atkinson, C. and Checkland, P. (1988) "Extending the metaphor 'system'". Human Relations. 41(10):709-724
Axley, S. (1984) "Managerial and organizational communication in terms of the conduit metaphor". Academy of Management Review. 9(3):428-437
Benford, S. et al. (1993) "From Rooms to Cyberspace: Models of Interaction in Large Virtual Computer Spaces". Interacting with Computers. 5(2): 217-237.
Brief and Motowidlo (19XX)
Burke, W. (1992) "Metaphors to consult by". Group and Organizational Management. 17(3):255-259
Burt, R. (1978) "Cohesion vs. Structural Equivalence as a Basis for Network Subgroups" Sociological Methods and Research. 7:123-212
Crider, C. and Cerillo, L. (1991) "Systems of interpretation and the function of metaphor". Journal for the Theory of Social Behavior. 21(2):171-195
DiFranco, A. (1995) My ID. New York: Righteous Babe Records
Granovetter, M. (1973) "The Strength of Weak Ties". American Journal of Sociology. 81:1287-1303
Hofstader, D. (19XX) Godel, Escher, Bach: An Eternal Golden Braid.
Johanssen, R. and Swigert, R. (1994) Upsizing the Individual in the Downsized Organization. Reading MA: Addison-Wesley
Knoke, D. and Kuklinski, D. (1982) Network Analysis. Newbery Park: Sage Publications.
Krantz, J. (1990) "Generative metaphor intervention: A new approach for working with systems divided by conflict and caught in defensive perception". Journal of Applied Behavioral Science. 26(2):241-243
Lakoff, G and Johnson, M. (1980) Metaphors We Live By. Chicago: University of Chicago Press
Langford, D. and Jones, C. (1994) "The Kitchen Interface: a Lateral Approach to GUI". SIGCHI Bulletin. 26:2:41-45
Lyman, P. (1995) "Is Using a Computer Like Driving a Car, Reading a Book, or Solving a Problem? The Computer as Machine, Text, and Culture". In M. Shields (ed.), Work and Technology in Higher Education. Hillsdale NJ: Lawrence Erlbaum Associates
Mark, G., Haake, J. and Streitz, N. (1996) "Hypermedia Structures and the Division of Labor in Meeting Room Collaboration" Proceedings of the 1996 Conference on Computer Supported Cooperative Work. New York: Association for Computing Machinery
Marshak, R. (1993) "Managing the Metaphors of Change" Organizational Dynamics. 22(1):44-56
Morgan, G. (1983). "More on Metaphor: Why We Cannot Control Tropes in Administrative Science". Administrative Science Quarterly. 28:601-607
Morgan, G. (1986) Images of Organization. Newbery Park: Sage Publications
Neilson, A. (1992) "A new metaphor for strategic fit: All that jazz". Leadership and Organization Development Journal. 13(5):3-6
Peffer, J. and Salancik, G. (1978) The External Control of Organizations. New York: Harper and Row
Peters, T. (1992) Liberation Management. New York: Alfred Knopf
Pinder, C. and V.W. Bourgeois (1982) "Controlling Tropes in Administrative Science". Administrative Science Quarterly. 27:641-652
Rice, R. and Richards, W. (1985) "An Overview of Network Analysis Methods and Programs" In B. Dervin and M. Voight, Progress in Communicxation Sciences. 6:105-165
Schall, M. (1983) "A communication-rules approach to organizational culture". Administrative Science Quarterly. 28(4):557-581
Shannon, C. and Weaver, E. (1949) The Mathematical Theory of Communication. Champaign: University of Illinois Press
Stone, A. (1995) The War of Desire and Technology at the Close of the Mechanical Age. Cambridge: MIT Press
Tolkein, J. (19XX) The Lord of the Rings.
Turkel, S. (1996)
Wasserman, S. and Faust, K. (1994) Social Network Analysis. Cambridge: Cambridge University Press
Weick, K. (1976) "Educational Organizations as Loosely Coupled Systems". Administrative Science Quarterly. 21:1-19
Williamson, O. (19XX)
Wozny, L. (1989) "The application of metaphor, analogy, and conceptual models in computer systems" Interacting with Computers. 1(3):273-283