In a previous special issue of this journal, David Boje and I (Luhman & Boje, 2001) attempted to address the question of what is complexity science. Our answer focused on the narrative research of organizational discourse. We stated that complexity science is a useful metaphor for understanding of an organization’s contextualized and emergent discourse as members interpret, re-interpret, negotiate, and re-negotiate discourse within a spatial/ temporal intersection. The study of organizational discourse emphasizes the collection of stories and drama (with characters, sequence of events, plots, and moral visions) as well as chronicles, but may include such items as simple gossip, official pronouncements, press releases, analytical reports, historical archives, or visual images. Narrative scholars generally make two assumptions: organizational members communicate primarily by story; and, an organization can be conceptualized as a storytelling community (Boje, 1991). The storyteller (or writer) and the story listener (or reader) are co-constructors of each organizational event. A multiplicity of stories gets enacted simultaneously in a multiplicity of sites (or texts), through numerous brief encounters (or quotes and citations), in and around organizations (or the printed product). According to Fisher (1984: 3) facts are more intelligible when combined into a good story to bind them together. Weick and Browning (1986), following Fisher (1985a, 1985b), suggest that as storytellers we utilize our told stories to determine, justify and guide our individual lives, and when we need to judge any facts we always question if they cohere against our own stories. Stories, themselves, adapt to give explanation to contingent events, such as founding tales (Clark, 1972; Martin, et al., 1983), mergers and acquisitions (Boje, 1991), or leader succession (Gephart, 1991). Stories also shape the identity of organizational members (Wilkins, 1983; Weick 1995). Martin, et al. (1980) describe how stories provide scripts that members use to organize their behaviors. Work by Michael White and David Epston (Epston, 1989; White, 1989, 1991; White & Epston, 1990) help people to re-story their individual lives, while Barry (1997; Barry & Elmes, 1997) and Thompson (1996) apply re-storing strategies to organizational change initiatives.

As humans we tell stories, and make drama, we attempt to make our narratives meaningful to the listener, to help them see connections and to partici-pate. In each telling, or dramatic performance, the narrative may change as we respond to the reactions ofparticipants. We may draw on other stories as com-parisons, embellishments, to situate our narrative in a broader discursive space, or orient the listener by linking our story to theirs. In other words, our narratives are ongoing linguistic formulations, composed in the moment, and responsive to the circumstances of a particular time and space. This is not necessarily a linear or a cyclical process, but a responsive one. As Bakhtin (1986) notes, meaning occurs in the interplay between people’s spontaneously responsive relations, to each other and the otherness of their surroundings. My contribution in this article is to advance an understanding of what I describe as ‘narrative processes’ within the metaphorical perspective of organizational discourse as a complex system. Our recent scholarship has given additional insight into this metaphor. I propose that there are three significant narrative processes, i.e., ways in which people in organizations story events:

  1. Story coercion which is derived from the concept of ‘micro-level hegemony’ (Boje, et al., 1999);

  2. Story weaving which is derived from the concept of ‘narrative temporality’ (Cunliffe, et al., 2004); and,

  3. Story betting which is derived from the concept of ‘antenarrative’ (Boje, 2001).

This article has three main sections: the discussion of the metaphor of organizational discourse as a complex system, the presentation of the three ‘narrative processes’ in organizational discourse, and a discussion of implications.

Organizational discourse as a complex system

Complexity science is understood as a set of presuppositions that shift science beyond deterministic and reductionist paradigms. This shift involves accepting the following presuppositions: life-systems are very complex and ever changing, thus, are very hard to model; any ignorance of the initial conditions of a life-system make any prediction impossible; order emerges out of chaos; and, irregularities emerge out of order. Complex systems, while unpredictable at the level of detail (e.g., the level of individual human behaviors in an organization), are predictable at the level of structure (e.g., the systemic processes of an organization as a whole). In the words of Peter Allen:

“The science of complex systems is about systems whose internal structure is not reducible to a mechanical system. In particular, it is about connected complex systems, for which the assumptions of average types and average interactions are not appropriate and are not made. Such systems coevolve with their environment, being ‘open’ to flows of energy, matter, and information across whatever boundaries we have chosen to define. These flows do not obey simple, fixed laws, but instead result from the internal ’sense making’ going on inside them, as experience, conjectures, and experiments are used to modify the interpretive frameworks within. Because of this, the behavior of the systems with which each system is coevolving is necessarily uncertain and creative, and is not best represented by some predictable, fixed trajectory” (Allen, 2001: 39-40).

Information in complex systems, as time moves forward, converges on, and then fluctuates apparently at random around, a space called an ‘attractor’. Attractors in complex systems are structures containing vast amounts of information on vast amounts of interlinking variables as well as on changes in patterns over time and through space. Attractors evolve out of the interaction of individual components within a complex system, and may even evolve out of a coherent effort of these individual components (Luhman & Boje, 2001). In complex systems there are four basic types of attractors (Allen, 2001):

  1. Point attractors – patterns ofbehaviors and processes that return to same state over time;

  2. Cyclical attractors – patterns of behaviors and processes that repeatedly travel within a defined space;

  3. Chaotic attractors – unstable behaviors and processes; and,

  4. Strange attractors – emergence over time of new patterns of behaviors and processes unpredictable from the original state.

It is with strange attractors that the complexity occurs with the ‘emergence’ of a new set of interacting factors that mutually support and complement each other (Allen, 2001). Emergence (see Hodgson, 2000, for a history of the concept) in complex systems exhibits the characteristics of radical novelty (not previously observed or predicted), coherence (maintained over time), globalness (seen only at the macro level), and nonlinearity (containing positive and negative feedback loops) (Goldstein, 1999).

The study of organizational discourse, as seen through the metaphor of complexity science, is an understanding that the whole is a unique entity that is never definitive, but ever emerging (Lefebvre & Letiche, 1999: 13). Organizational discourse flows through time and space. As time moves forward, a new system of organizational discourse emerges creating a different collectively constructed discourse of organizational reality (Luhman & Boje, 2001). Discourse is defined as verbal or written expressions and discussions on a related topic, and also a process or power of reasoning (American Heritage College Dictionary, 1993). This basic definition brings us to the modeling of communication in the field of organization studies as per the Shannon-Weaver-Schramm model (Shannon, 1948; Weaver, 1949; Shannon & Weaver, 1949; Schramm, 1954). In order to have discourse, communication must occur which is defined as “an attempt to create common meaning or understanding between the sender and receiver of a message. The message can be distorted by noise from a variety of sources and thereby interfere with the attempt” (Church, 1999: 4). Communication is the sending and receiving of information as well as the proper interpretation of meaning through common ‘fields of experiences’ and effective feedback.

This leads to a probability function of interpretation (Shannon & Weaver, 1949: 8-16) as “the probabilities derived from choice rather than ignorance; they reflect how probable it is when we would choose one message element rather than another, given a known ensemble” (Hayles, 1990: 54).

A simple case is represented in Figure 1. In this situation (see Shannon & Weaver, 1949: 15), the choice here is between two possible messages where one choice is represented asp1 and the other is p2 = 1 – p1. When the two messages are equally probable, that is when p1 = p2 = 0.5, then information is at its highest level. When one message is more probable than the other, that is when p1 > p2 or p2 > p1, then information is decreasing. And when one message is very probable over the other, that is when p1 nears 1 or p2 nears 0, then information is at a very low level.


Fig. 1: Information sharing and meaning making as a probability function

Source: Hayles, 1990:53

From this representation, discourse within an organization is seen as a probability function of a field of choices in which meaning takes place. Accord-ingly, organizational discourse influences the field of choices as a primary source ofmeaning making. With this discussion of the metaphor of organizational dis-course as a complex system, I move now to present the three proposed ‘narrative processes’ in organizational discourse.

Three narrative processes

Recent research with my colleagues has brought forth the idea that there are the three ‘narrative processes’ in organizational discourse. First, the conscious or unconscious effort of individuals to create universal meaning over organizational events is story coercion. Second, since narratives are always open to the interpretations of participants, story weaving takes place in the context and in the point of enactment. And third, when potential storytellers attempt to pre-story organizational events, they are story betting.

First narrative process: Story coercion

The first narrative process of story coercion is derived from the Boje, et al. (1999) concept of‘micro-level hegemony’ of individual narratives. Micro-level hegemony is the conscious or unconscious effort of individuals to create and establish meaning over organizational events. Individual narratives are a selective seeing which benefits some over others – a struggle of the individual to actively re-story meaning from one event to the next. Those organizational actors with power have more opportunity to maintain their version of events. Organizations are ‘associations’ or ‘polities’ where power and legitimate authority over member’s behaviors and decision-making is distributed (Putterman, 1988), and where the “relations of meaning, as well as relations of production are central to the structure and functioning of organizations” (Clegg, 1989: 112). Power is not established through position or structure but rather is “a set of strategic practices reproducing or transforming a complex ensemble of relations” (Clegg, 1989: 111). Thus, story coercion is the power of individuals to tell stories that inscribes or constrain individual action. An example from the Boje, et al. (1999: 349) study tells about how members of the Choral Company (a small singing group) kept secrets amongst themselves, and from outsiders, to maintain a public face of group harmony.

“The Choral Company (CC) maintained secret stories as well. First, the day of the concert at the academy meeting I (Baack) was told by Jon that a member of the group had booked the plane flight to New Orleans for the wrong day. Two days before the performance, the group was faced with the possibility of canceling the trip if an alternative flight plan could not be found. The discovery of the booking error had generated considerable animosity between some of the members, obscenities were shouted, and fingers were pointed. Yet, the CC resolutely kept this conflict from being shared with (Boje) and the academy audience.

At the meeting, members of CC knew that Marcus, the piano player, was leaving the group. His frustration was caused by the romantic overtures of another member through phone calls and visits, and he simply was not interested. He finally had notified the group that he was leaving in order to ‘go back to school’ (he took one class the next semester). Some members knew the real reason for his departure, whereas others did not.

Even I (Baack) was somewhat frustrated with the group because they had divided funds they had raised for the trip among the eight nonacademic members, leaving me to pay my own way, using the rationalization that I was receiving ’school’ money. I eventually complained to Jon about not only the funding issue, but the fact that only three CC members (Jon, Meg, and Jae) ever actually got around to thanking me for scheduling the event in the first place. The story being told by members of CC was about harmony and fine-tuning that harmony; I had willingly played along, all the while knowing of these secret stories.”

Organizational actors have an active will to power that Nietzsche views as a series of moments, each of which is related to what came before it and what will follow it by the intentions of the actors on the scene at that time (White, 1973: 363). Stories pass through much reinterpretation with each new event. The storytelling system does its work in the struggle between master and local, dominant and marginal narratives (Boje, 1991, 1995). The organization is storying and re-storying itself, constructing and re-constructing the scripted roles its members will act out. It is a play in which some players get bit parts while others are treated with full humanity. And some players have more rights and power in making up stories than others. It is not whose story is best, others’ or mine, “but who has the power to make his story stick as one that others will choose to live by or in” (White, 1987: 167).

Second narrative process: Story weaving

The second narrative process of story weaving is derived from the concept of ‘narrative temporality’ that was first presented in Cunliffe, et al. (2004). In this article, a new set of presuppositions was presented on the idea of time. Based on two amendments to Paul Ricoeur’s work in Time and Narrative (1984), we re-storied narrative research in organizations as a fluid, dynamic, yet rigorous process open to the interpretations of its participants and situated in the context and point of enactment – a process of story weaving.

The first amendment dealt with the presupposition that organizational discourses are not already constructed narratives, rather narratives are perfor-mances in the moment, “a product of imaginative construction” (Mink, 1978: 145). Life is lived in the moment and much of our sense making also occurs in the moment. Even though narrative knowledge is about meaningful time, the performance of narratives takes place in practical circumstances (contexts, spaces) and in particular moments (time) in which meanings may vary. Thus, the process of narrating is crucial to meaning making: what I say, how I say it, what the listener hears, how s/he feels, how s/he reacts or responds. We may extend this argument to suggest that space and time are not necessarily separate dimensions because the unique circumstances of each moment, the context of performance and interpretation, and the specific interrelationships and connections that occur in the moment all interweave to create a unique discursive space (Cunliffe, et al., 2004). In narrating our experiences we engage in a relationally responsive activity as we attempt to make our narratives meaningful to listeners in a particular context, to help them see connections and to participate (Cunliffe, 2002). Shared discursive spaces emerge because we live in communities of practice (van Maanen, 1998) and draw on other stories (collective or individual) as comparisons and embellishments to situate our narrative in a broader discursive space or to orient the listener by linking our story to theirs.

The second amendment dealt with the presupposition of perpetual referring within Ricoeur’s ‘threefold mimesis’, which occurs across time and contexts resulting in multiple threads of earlier narratives weaving together into multiple present ‘emplotments’, and continually recreating multiple future emplotments. This amendment incorporates the reflective consciousness of temporality from Sartre’s Being and Nothingness (1956). Sartre (1956: 130) distinguishes between a static linear temporality and a temporal multiplicity. In the former, time is irreversible; narrators narrate order in terms of chronology, before and after. In the latter, time is dynamic which does not separate into discrete units located before or after events, but is experienced as an infinite dispersion of multiple pasts and futures. This interpenetrating of present-past-future is experienced through a unity of perpetual referring – a process of reflection reflecting. So we reflect on past events, but that reflection is influenced by both our currently experienced moments in time and the future moments we may be anticipating. It is within the moment of storytelling that the circle of mimesis occurs as we combine time (past, present, future), stories (memories, attention, expectations), and meaning to shape actions and identities in conscious and unconscious ways (Cunliffe, et al., 2004).

An example of story weaving from the Cunliffe, et al. (2004: 274) research is when one of the authors (‘A’) in a conversation with a proj ect manager (‘P’) create meaning together:

P: “We tend to do a lot of that around here where we like the idea of having an expedient answer and fail to understand the reason – which then doesn’t get communicated back to the rest of the organization.”

A: “Is that because there are no clear structures for dealing with this?”

P: “Yeeess. I was talking to the IT Manager this morning … and I said, A month ago we were talking about how long it would take us to get the detailed layouts for this group, and a month ago I would have said two weeks. And yesterday in our meeting you said, ‘I told you it would be four weeks’ and I said ‘Yeah ’. ’ Because what I’ve realized is the ratio of managers to people is so high for this group… who have to agree to every scheme and it takes time…”

A: “So is it the sheer number of people – and trying to coordinate all their inputs…?”

P: “Yes, and some of it is because there isn’t any kind of authority scheme that would help reduce those numbers. The other thing I came across recently was…”

[Later I ask:]

A: “So how do you feel about this unpredictability and uncertainty?

This narrative example demonstrates how both researcher and participant drew on past narrations (organizational and academic) to story weave in the moment and anticipate future narratives about how they might write up the research. Your reading, in a different time and space, may draw on other nar-ratives to make sense, agree, disconfirm, or create new narrations (Cunliffe, et al., 2004).

Third narrative process: Story betting

Finally, the third narrative process of story betting is derived from the concept of ‘antenarrative’ as developed by Boje (2001). Antenarrative has a double meaning: first, as being before, and second, as a bet. A narrative is an account of incidents or events that adds plot and tighter coherence to a story line. Antenarrative is therefore ‘ante’ to the narrative; it is in advance of the narrative. ‘Ante’ is also a bet, something to do with gambling and speculation. Thus, an antenarrative is a ‘pre-story bet’. It has not yet become realized in the world. Antenarratives depend upon an expanding network of actors to work in bits of context, to send the antenarrative along its way to becoming a story (Boje, 2001). Here is an example of story betting from Boje’s (1991: 115) storytelling research of an office supply firm. The attempt to bet on a version of events is clearly seen here as Sam plots how Raymond ran things in the past.

Sam: “I think five years ago there was no Executive Committee. He just ran the place the way he wanted as if you wasn’t here. Raymond was [conglomerate’s] man and he did what he pleased. I guess you heard this all already

Dave: “I heard about the high growth” (nodding)

Sam: “Yes we picked up San Diego”

Dave: “Reno?

Sam: “Ed Fox picked up Reno and I guess you heard the whole story before. That’s why we picked up Reno because Ed Fox had that with his father. You know.”

Every once in a while, someone interrupts the flow of experience and asks you to give an account of what is going on. Your mind races, experiences come to mind, a plot thickens, and you begin to speak, and a ‘pre-story’ is told. You are living experience before narrating it, before someone requires that you provide a story with the coherence of a beginning, middle, and end. And then it is out there, but you know it is only one way of beginning to create a story. Others will have their ways because the collective memory is always being reworked and never completed (Boje, 2001).

The three narrative processes in organizational discourse, i.e., ways in which people in organizations story events, are story coercion as the conscious or unconscious efforts of individuals to create universal meaning over organizational events; story weaving since narratives are always open to the interpretations of participants while being situated in the context and point of enactment; and, story betting when potential storytellers attempt to pre-story organizational events. Thus, organizational narratives are never definitive, but always emerging as they flow through time allowing for their re-interpretation and re-negotiation. Most narrative researchers assume that language helps us decipher already-made significations with relatively fixed meanings, so we can study the general language systems, structures, stories, etc., to see how subjects construct their world, and then theorize about their discursive communities. The three narrative processes suggest meaning takes place in the telling and listening of narratives, therefore, there is not one self-contained narrative to analyze.


One implication is the need to develop more participative forms of narrative research where scholars try to discover meaning in the moment and where all research participants are heard. If the metaphor of organizational discourse as a complex system is accepted, then there is a need to explore different stories and interpretations as a means of narrating lived experience, and accept that each act of narrative interpretation is unique. Because participants in organizational life continually engage in narrative performances, and because we are not all-knowing researchers, we cannot precisely explain organizational events. What we can do is collect narratives of organizational events, participate in the sense making processes, and offer potential interpretations.

One way is to make organizational discourse research ‘polyphonic’ is by having others participate as voices in the narrating of the research. The ideal for polyphonic data gathering techniques is best represented in Peter Reason’s (1994) discussion ofthree participatory research strategies. Reason “sees human beings as co-creating their reality through participation: through their experience, their imagination and intuition, their thinking and their action” (1994: 324). Generally, the practice of discourse research involves watching, listening, asking, or reading by the scholar to create a credible interpretation of the stories of social actors engaged in organizational activities. The goal here is to shift the power of ‘reality creation’ away from the scholar toward those being studied.

There are three participatory research strategies that Reason (1994) attempts to integrate: cooperative inquiry, participatory action research, and action inquiry. Cooperative inquiry is a participatory strategy that focuses on serving a group of action-oriented researchers. Participatory action research is a strategy that focuses on serving the community by developing skills and empowering the social actors affected by the research project. Action inquiry is a participatory strategy that focuses developing the skills required to conduct participatory research. The particular techniques used within each of these strategies that would make the narrative data collection process polyphonic include engaging participants to create a dialogue between the scholar and participants, and then reflecting on the beliefs, observations, and recordings of experiences, to emphasize differences between the scholar’s stated theory of action and the theory-in-use. Participation by the social actors under study, and the scholar’s self-reflection over the data collection process, should allow for voices to be heard that are usually silenced.

A second way is to help social actors to ‘restory’ their narrative organizational life through the technique of deconstruction. Deconstruction problematizes taken-for-granted ways of being and knowing, so-called truths, that are objectified and reified as the processes of the construction, discipline, and recruitment become hidden or forgotten. “To deconstruct is to actually analyze the relations between the dualities in stories – such as the positive and negative, the central and the marginal, the essential and the inessential, the insider and the outsider – to show the ambiguity embedded in them and to show the storytelling practices used to discipline particular meanings” (Boje, 1995: 1007). Deconstruction is a critical reading of the ‘self’ and the ‘organization’ as a narrative script given by global and societal storytellers to habituate docile behavior in consumer, capitalist, educational and government enterprises. The analysis of how one perspective dominates another is re-storied to disentangle the forces acting upon the narrative of an organizational self. The use of narrative power to keep the individual a docile, rather than a pro-active agent, is deconstructed. Deconstruction facilitates in questioning the narratives organizational actors are recruited to play and that shape their lives. In re-storying, the individual might confront the narratives they accepted to play out, and live a story that is more dignifying to them.

Organizational research is usually an attempt to find ‘a’ reality of effectiveness and legitimation that can be transferred onto other social situations. Alvesson and Willmott (1992: 454-460), taking a critical theory approach to research, suggest a three-step methodology to avoid this destructive tendency of organizational scholars:

Step 1: Listen to People. Conduct studies that capture the sense making process of everyday practical life. The goal being to discover the stories of values and belief that influence the social reality viewed by each individual organizational actor.

Step 2: Deconstruct Dualities. Find the dualities (the dichotomies of beliefs and values) that exist in organization stories, and their hierarchical orders (which term in the duality dominates the other). Find the hidden, marginalized term or voice and attempt to demonstrate what a reversal in the hierarchical order would be like, attempt to let the silent voices be heard.

Step 3: Allow People to Re-Write Their Stories. By allowing organizational actors to self-discover the stories that guide their everyday, practical life, to view how a reversal of dominant to subordinate values and beliefs may look like, researchers might engage each individual in their own liberation toward autonomy.

The key is to attempt an evoking of discourse and not a simple representation of it; to move away from mimesis and back to experience where the organizational actor is reintegrated into society and life (Tyler, 1986). From this foundation, narrative researchers might seek to understand “why individuals, whose lives are far from any point of control in organizations, usually accept and willingly fulfill the ordered lives that tradition, their jobs, and the organization prescribe for them” (Greenfield, 1979: 107). Stories discovered in this quest might enable individuals within the organization to view the social constructions that dominate her or his life.