Printer Friendly

Collective Form: An Exploration of Large-Group Writing 1998 Outstanding Researcher Lecture.

Whether a collective mind forms in large-group writing in the workplace is the focus of this article originally given as the 1998 ABC Outstanding Researcher Lecture. This article is based on a five-year ethnographic study that describes and analyzes a three-month group writing process that created a computer service-level agreement, involving a 20-person cross-functional core more than 100 other collaborators at a major corporation. The article discusses "collective form" in two senses: First, a document's evolving form or superstructure produced a collective schema that allowed the group through a process of equilibration (Piaget, 1981) to adopt outsider boilerplate into a more situated general model and then into a situated document. Second, architectural forms motivated and molded group activity in several ways. To combat group apathy, the leaders appropriated an in-demand meeting room for the project, positioning the project as high-status in the center of the workflow. Group leaders prominently displayed a task completion check-off chart that, in a downsizing environment, helped both to coordinate group activity and to encourage completion.

Keywords: Collaborative Writing, Collective Mind, Cross-functional Groups, Large Group, Team Management.

Trends ends in business and government affect the very nature of writing in those environments, as Tebeaux has noted (1996). One of the most widespread and genuinely altering trends, brought on by global competition, is the move in the 1980's and 1990's of U.S. companies from taskfocused to process-focused management, often through the managerial approaches of total quality management and/or reengineering. For example, the tasks involved in order fulfillment, such as taking the order, entering it into a database, checking the customer's credit, and so forth, are not in isolation valuable to the customer. Yet before process-focused management (and even today in too many cases), management focused only upon how many orders were taken, how many credit checks were conducted, and so forth, in isolation from the rest of the process. But only when these tasks are integrated do they create value (Hammer, 1996, p. 5).

The continuous improvement or fundamental redesign of business processes in organizations requires much writing, first, to identify these complex processes; second, to design incremental or radical improvements; and third, to implement the changes. Because these business processes typically span many departments in a company, their documentation and adjustment or reinvention often creates the need for a large-group writing effort. Ede and Lunsford have defined group writing as "any writing done in collaboration with one or more persons" (1990, pp. 15-16). Large-group dynamics begin when the number of group members exceeds 15-20, as Gilmore and Barnett note (1992, p. 539).

Although researchers have studied many small collaborative writing groups, only two studies have focused upon any kind of large group collaborative writing (Mclsaac & Aschauer, 1990; Kleimann, 1989). Mclsaac and Aschauer discuss other important topics (e.g., writing in engineering and persuasive strategies) as much as they do large-group collaboration. Kleimann (1989) focuses upon the review of documents by large groups in the workplace. Thus no one has described in depth a collaboration in which a large group had input into the planning and/or writing that went into the first complete draft of a document. Beyond that, no study has described the training of leaders and their facilitation of concurrent collaborative subgroups working on a large document. Noting this incomplete information on large-group writing, Association of Business Communication Outstanding Researcher Janis Forman (1991, p. 247) called for more research in this area. This study begins to fill that gap.

This article comes from a larger ethnographic study describing and analyzing a three-month group writing process involving a 20-person cross-functional core group and over 100 more than collaborators at Montmanche Corporation (pseudonym), a major corporation.

Overview and Conceptual Framework

My study investigates how an Information Technology department (IT; a department providing computer support) at Montmanche Technological Services (MTS) orchestrated the writing of a service level agreement (SLA) in order to survive after a narrow escape from the "outsourcing" axe. Initially, project members had difficulty forming a functional group.

This article describes two reasons the group ultimately coalesced.

The exigence of the project occurred in 1994. In return for continued funding, Montmanche Technological Services agreed to provide among other things a set level of computing service to users in the rest of the company (for example, all help calls answered in four minutes). Before the SLA project, IT was not as accountable-for many IT services, departments paid a flat rate and had to put up with what they got. On the other hand, departments had found ways to subvert existing procedures to overload MTS, creating processing bottlenecks at the end of the day, month, quarter, or year. But now processes were to be defined in a Service Level Agreement to produce more measurable indices, per-transaction billing, and fewer bottlenecks. The resulting document would help both MTS and the departments reach optimal levels of service and cost.

The document was clearly needed, and upper and middle managers defined it without incident. But after it was delegated to a project team composed chiefly of MTS departmental managers, for the first few months the writing group's project leader struggled unsuccessfully to get most group members engaged. As the first project leader left to take over another department, her replacement got the writing of the document put into the job objectives of the project core group members. Now their performance with the group would be measured and used in decisions to downsize or to award raises and promotions. The result of this action was that all members began to participate, although many held different mental models of the document. So on January 13, 1995, a Friday appropriately enough as it turned out, the core group met to develop a clear understanding of what they were trying to create.

Developing this understanding was crucial. As Weick and Roberts (1993) note, for a group to serve its purpose, a collective mind must form. Weick and Roberts define "collective mind" as a pattern of heedful interrelations of actions in a social system (p. 357). The output of the functional collective mind is not "groupthink," a group's putting such a high premium on harmony that it punishes dissent and ignores bad news from its outside environment. Instead, a collective mind is found in the heedful interrelation of group members. People are heedful when they are critical, careful, consistent, purposeful, vigilant, and conscientious (p. 361). Asch (1952) describes three key components of the heedful interrelation of the collective mind:

(1) Members contribute their actions while envisioning (representing to themselves) a system of joint actions, and members subordinate their contributions to the system of joint interactions they envision. That is, they act in consideration of and accompaniment to other group members' actions on the same group project.

(2) A joint system of interaction is indeed produced by members' envisioning pertinent parts of the system of interactions and subordinating their contributions to it. This system or process has a definite form or pattern that is larger than the individual. As Asch notes,

Such a system does not reside in the individuals taken separately, though each individual contributes to it; nor does it reside outside them; it is present in the interrelations between the activities of the individuals. (1952, p. 252)

(3) The outcome produced varies as a function of the degree of heedfulness and of strength (loose interrelation or tight interrelation) that binds the activities together. Certainly, other conditions that Asch does not mention, such as inclement weather or terrorism, may affect the outcome of the group; nevertheless, the degree of heedfulness and strength are critical elements in a successful outcome.

Whether a collective mind can form, however, has been questioned by Newell (1990, pp. 490.491), who argues that individuals' long-term memories contain a lifetime of information, yet groups last relatively briefly. Newell sees any collective mind requiring the combination of these lifetimes of memory. But the sum of all the knowledge of all individuals in the group is not needed to still have the heedful interaction of a collective mind (Smith, 1994, pp. 101-102). Newell also argues that individuals cannot communicate all pertinent information to the task in the time allowed by meeting. But as Smith notes, individual minds operating alone do not always use all the potentially pertinent information they themselves have for each mental operation performed (pp. 101-102). Beyond that, groups do not have to form a collective mind entirely in "real time." Thinking about a document by a group can develop in part outside of group activities and aggregatively, yet this thinking can be heedful of the group's system of joint actions. One payoff of heedful interrelating during and apart from group activities is that such heedfulness can activate sufficient reserves of individuals' knowledge so that the group can understand unanticipated events that quickly develop in unexpected directions (Weick & Roberts, 1993, p. 366).

Groups do not necessarily form a collective mind, however, as Weick and Roberts note. And an established collective mind can break down if significant loss of heed in interaction occurs (p. 373). When interrelating breaks down, individuals represent in less detail to themselves co-workers in the system, individuals shape their contributions less by what they anticipate other workers' responses will be, and individuals draw their mental boundaries of the system more narrowly, with the result that their efforts to subordinate their actions to the system fail (p. 366). Interaction becomes careless: The group loses its intelligence; individuals may still be heedful, but not in relation to others (p. 366). As the number of interactions among group members decreases, the collective mind simplifies and soon dissolves into individualism. This individualism increases the project's vulnerability to error because group projects rely upon the interrelated labor and knowledge of individuals (p. 378).

Such a group "loss of mind" occurred at the core group's document planning meeting on January 13, 1995. The group's new leaders intended the session to "get everyone on the same page," sharing the same understanding of the SLA document and of the process to create it. But many core group members had not read the draft of the SLA and did not have it with them in the meeting. Instead of an orientation, a confusing 90-minute document-planning session occurred. When I identified and traced the topics brought up in the meeting, the most telling statistic regarded procedural topics, topics that comment upon the way that the meeting is run, rather than on the task the meeting is supposed to accomplish. Nearly thee-fourths of procedural topics were mentioned only once, suggesting a great number of different attempts to get the meeting under control. Note the even distribution of procedural topics across the whole meeting (see Figure 1).

After this "mental breakdown," the group panicked. The next day, members tried to get permission to abandon the assignment, then settled for taking orders from the project manager rather than having to engage in more peer collaboration. After this "revolution" against consensual leadership, the facilitators designed the document, assigned everyone tasks, trained them to complete them, and oversaw the completion of the document, during which time some of the objectives for the project were reduced. The group overcame its weak beginning and got the document done under a tight deadline because of its heedful interrelation. This article examines two ways that a collective mind was "formed": (a) the group leaders established a common form that facilitated the heedful interrelation of group members in assimilating local knowledge and accommodating this form to meet the needs of Montmanche; (b) architectural forms motivated and molded group activity.

Research Methods

Early in this qualitative, ethnographic study I made my understandings of the subject as explicit as I could in a conceptual framework and then successively revised this framework to account for what I found in the field.

I spent 480 hours at the site in full-time participant-observation from January 5 though March 31, 1995. During this time, I

* Observed and taped 48 collaborative sessions.

* Conducted and taped 64 open-ended and discourse-based interviews (Doheny-Farina, 1984).

* Collected personal documents, internal official documents, and external official communication.

* Recorded observations in fieldnotes and a process log.

To analyze these materials, I

* Derived analytic categories and codes (Miles & Huberman, 1994) from fieldnotes and conceptual framework.

* Coded 1,500 pages of audiotape transcriptions and 500 pages of fieldnotes and documents.

* Established a database and entered data for a few documents, then supervised assistants Tony Baker and Vicky Cummings' document coding and data entry. More than 1,800 editorial changes in document drafts were identified and entered into the database matrix (Cross, 1994, pp. 159-160). Data entered included

* All editorial changes made to documents.

* Kind of change (e.g., insertion, deletion, replace, join, split).

* Rhetorical/grammatical description of each change (e.g., clarity, conciseness; Cross, 1994, pp. 159-160).

* Editor's explanation of the change.

* My explanation for the changes (in analytic codes).

* Medium used to communicate revision advice.

* Cross reference to the changer's explanation in transcriptions.

* Conducted database/textual analysis using matrix, drafts, fieldnotes. [1]

* Drafted representation.

To authenticate my findings, I

* Spent a prolonged period at the site.

* Attained multiple time perspectives (Doheny-Farina and Odell, 1985).

* Compared several types of data.

* Compared the perspectives of numerous informants.

* Received feedback on my conclusions from book and article reviewers and by presenting various findings at six national or international professional conferences.

A Common Form

A collective mind coalesced during the process of adapting a service level agreement (SLA) from another company (which I will call Instapolis Corporation) to fit the situation, purpose, and audience at Montmanche. The Instapolis SLA was chosen by the initial project leader from among several SLAs given to her by a consultant. This SLA is basically a legal contract between the data center and its "customers," stipulating levels of service provided and roles and responsibilities of both parties (see its table of contents in Figure 2). This SLA does not differentiate its customers into their departments (business units) and describe their missions and software applications. Instead, it lumps all departments into "users." The SLA devotes its space to carefully defining different measures of service; procedures of review, crisis negotiation, change, and renegotiation; computer responsibilities of both parties; and disaster recovery.

From the first Montmanche draft that imitated the Instapolis SLA contract, the Service Level Agreement became progressively differentiated into three sections. The section I will discuss, Section A, became 22 documents, almost every one of which described a different business unit of Montmanche. Each SLA contained among other things a new large section and appendix whose purpose was to educate MTS about the business unit in the agreement. The group's progressive differentiation of the document reflects the group's psychological process of equilibration, a balance of two mental processes identified by Piaget (1981). The first mental process, assimilation, is the reshaping of the mental representation of the currently examined object to fit the perceiver's pre-existing cognitive structures, in this case, the pre-existing perceived forms of the document for Montmanche.

The concurrent process with assimilation is accommodation, in which the perceiver's pre-existing cognitive structures change to accommodate novel aspects of what is just now being perceived. This action is not fully discrete from accommodation:

Assimilation and accommodation are not two separate functions but the two functional poles, set in opposition to each other, of any adaptation. So it is only by abstraction that one can speak of assimilation alone, ... as constituting a function of essential importance; but it must always be remembered that there be no assimilation of anything into the organism or its functioning without a corresponding accommodation and without such assimilation's becoming part of an adaptation context. (Piaget, 1971, p. 173)

The way equilibration worked at Montmanche was as follows: The Initial Project Leader began the drafting stage by assimilating the Instapolis SLA into her pre-existing knowledge of the SLA by making some changes to the Instapolis document to suit Montmanche's situation, purpose, and audience, but still closely imitating the Instapolis SLA. This document was circulated to all other core project group members. Rick Frey (pseudonym), a core group member who had been assigned to help with the initial drafting of the document, considered the first SLA draft and revised it to accommodate his novel perception of the rhetorical situation at Montmanche. Then the Assistant Project leader Nancy Durnil (pseudonym) revised that draft to accommodate her novel perceptions of the rhetorical situation. The document later changed to accommodate perceptions of a pilot department (Consolidated Insurance Financial) and then to accommodate descriptions of 21 departments in the company. Alter it was carefully read by all group mem bers--after the Christmas and New Year's holidays and after the failed planning meeting--the document became the chief locus of group interaction because members did not work on the SLA together all the time. Yet when individual members worked on it, they did so heedfully, envisioning the other group members' activities and subordinating their own actions to group goals and directives. Carefully reading the SLA document allowed group members to form a reasonably common schema--that is, an internal figurative representation--of the SLA. The common schema produced by the reading of this document allowed assimilation and accommodation to occur as the group developed its document.

Piaget notes that schemata must form in order for cognition to function. During the failed document planning session January 13th, the schema of the new draft of the SLA was not present in the collective mind, though it was in two or three individuals' minds. There was no shared schema in part because of a lack of heedfulness--few members had the draft that had been handed out two days before fresh in their minds, and no copies were distributed or referred to in the meeting. Group members generated new information including alternative concepts of the SLA expressed in metaphors, but the SLA group was unable to conserve anything or transform anything with the new information. In short, the group was unable to function. However, once new project leader Nancy Durnil assumed hierarchical leadership of the group, she created an annotated model SLA document and went over it section-by-section with group members who were to lead writing subgroups to produce a Section A of the document. She also gave them instructio ns on how to lead these collaborations with each Montmanche department. These members were then able cognitively to transform the model into integrated yet situated documents. The Section A's were integrated because there was continuity of form among their separate documents that allowed them each to function as part of the SLA. On the other hand, each Section A document was sufficiently situated to be approved by the department it described. Because subgroup leaders finally had a common form of the document in mind, they were energized by the business unit information they now had a way to process as group members. The result was esprit de corps, and it is important to remember that the French word "esprit" means both "spirit" and "mind."

Recapitulating this process suggests how a collective mind can function in a large-scale writing process. The writing of the SLA progressed from the close imitation of a legal model in the first draft to the transformation of its form to fit a new situation, purpose, and audience. In this process, the outside model served both as an initial form and as a mnemonic prompt, helping the collaborative group to think more specifically about the rhetorical situation for the task. The outside model clearly was not something that could be appropriated as is. In The Variables of Composition, Broadhead and Freed (1986) show that this was also the case with one of their subjects, who said he went into a proposal-writing situation thinking some boilerplate would work perfectly and came out with an almost completely different document. The writer said that all the boilerplate supplied was a few "neat ideas" (p. 58).

For MTS, immersing itself in the outside model though direct imitation was a start to thinking about the construct of the SLA. To move from the outside model to a situated document, MTS used the following process (see Figure 3).

The first draft of the Montmanche SLA (Document 2 in Figure 3) was closer to the imitated outsider SLA than it was to the final Montmanche product, and had no description of the functional areas. Three weeks later, the document became a generic model (see Figure 4), its particulars overtly fabricated, just providing predicated topics in the order they were to be addressed. The generic model still resembled the Instapolis SLA in stipulating levels of performance and assessment procedures. Next, the project leaders began constructing a prototype of the document with one of the business units serving as a guinea pig, Consolidated Insurance Financial (CIF). After more feedback, once Nancy and her assistant Rick designed, dispatched, and considered the results of a questionnaire eliciting information about the business units, Section A changed radically to describe the business unit's work, critical periods, and critical factors in getting the work done. The document had changed about 180 degrees from informing C IF about MTS (in the form of agreements, expectations, and policies) to informing MTS about CIF (describing CIF subunits and their needs).

Nancy and Rick wanted to use the prototype as a model for their team-leader training sessions. But although the previous generic model could be accessed by the whole corporation, the situated prototype ostensibly just addressed CIF (a document is situated when it appropriately describes and/or relates to its context).

To be sure that each MTS-led writing group in each business unit would create the appropriate section, Nancy and Rick wrote in the margins of the CIF model boxed labels indicating what should go where with arrows connecting them to textual features (see Figure 5). Nancy and Rick then trained their 20 core-group members to lead the business-unit collaborations. Using this prototype with the added overlay of explicit formal rules, each business-unit writing group wrote its own document (e.g., see Figure 6-one SLA Section A). Like the prototype, these documents did not include legal agreements but rather described many aspects of the business units, including their computer needs.

In resituating the SLA, the writing group went from imitating a specialized document from another company to abstracting what was valuable from that document for the Montmanche context--finding the topics and format that would fit the situation, purpose, audience, and then filling in the template with specifics. They had to travel up then down the ladder of abstraction to be able to make a situated document from another place into a situated document in a new context.

The progression of the SLA document can be explained by research by Brewer (1988, p. 17), who said that individuals tend to use the most specific level of abstraction that is available and adequate for effective categorization. The Instapolis SLA was a concrete document that was available. Rick found it wasn't adequate and changed it. Nancy revised it more. CIF department computer science representatives had Rick change it again to fit the CIF context adequately. To that prototype was added an overlay of abstract boxed labels with arrows connecting them to specific features so that the 23 other business units could use the structure to convey their local knowledge.

The transformational process from situated document to generic model to situated model with formal rules to new situated documents occurred because group leaders had to think abstractly-draw up a blueprint or outline-to find the aspects generalizable to another situation. Once the abstraction occurred, assimilation and accommodation could progress because the group had a way of arraying and absorbing the business unit's information and because the group could add to or modify the essence of the original (e.g., the November 1 draft) in response to the local (e.g., CIF's) situation, purpose and audience. The up-then-down progression seems to be the cognitive bridge to resituate the form. Neither the individuals nor the group may have been able to build a flat bridge from the close imitation of the outside document (Document 2 in Figure 3) to the situated prototype (Document in Figure 3). And they may not have very effectively begun the writing process with the vagueness of the generic model (Document 3 in Figu re 3) and moved directly into the situated prototype (Document 4 in Figure 3). The group leaders appeared to need the point of reference that was the close imitation of the outside model (Doc. 2 in Figure 3) before they could abstract appropriate topics. And they needed to burn the form of the SLA into the consciousnesses of the other subgroup leaders through meticulous training for the 120-member group to complete the document through heedful assimilation and accommodation.

Architectural Formation

Architectural forms also helped bring into presence and shape the collective mind, functioning symbolically and logistically in vital ways to influence the project.

Symbolic Capital

Architects in pursuit of the discretionary dollars of wealth emphasize what Harvey calls "symbolic capital," the accumulation of status objects that conveys the owner's taste and distinction (Harvey, 1989, p. 77). As Herb Hildebrandt noted in his Outstanding Researcher lecture in 1995, in commercial buildings with high levels of comfort and aesthetics, symbolic "funds" are designed to be expended on persuasive transactions. The financial capital put into aesthetics is justified by the expectation of attracting more of the desired kind of customers and employees.

Montmanche had expended considerable financial capital on aesthetics. When one passed the elaborate security, entered the Montmanche Building, and began to cross the marble parquet in the atrium, reflections ebbing like waves on the brightly polished floor, looking up to see the marble stair spiraling vine-like to the skylight, one experienced distinction. One felt valuable. This high aesthetic standard was maintained in its art gallery/corridors, elevator lobbies, corporate restaurant, and other areas, while a high level of comfort was maintained in the bright carrels, "all-day chairs," and other furnishings.

The sense of value of this experience was accentuated by the threat of its disappearance, recalled by the visual reminders of velcro-backed name holders on the carrels, temporary workers, security guards, and the knowledge that admission to this distinction relied upon an activated Montmanche pass card secured by being in the good graces of management in a downsizing and outsourcing environment. This symbolic context encouraged participation in projects most important to management.

Nonverbal Language

A human-made physical setting, as Becker notes, promotes power relationships we suppress or ignore in verbal communication (1981, pp. 99-100). In an open office such as the research site, what was considered critical was often indicated by its proximity to "prime real estate." Three senior managers were carreled in an area known in the culture as "the City," a namesake of London's financial square mile, with secretaries and subordinates surrounding them. With its "Race-Board" that had different cardboard automobiles representing project groups, The City was the center of managerial control and the focal point of action in the department.

Nancy's initial low-visibility, low-status location in "the Country" worked against her by marginalizing the project, but she and Rick convinced the Vice President to let them appropriate a glassed-in conference room ("the War Room") near the only entrance to the department, located on a short, central corridor that opened into the center of communication and power. Most people passed by the War Room several times a day. As managerial theory notes, controlling and changing the "stage," that is, the physical surroundings, helps control the action of organizational life. Appropriating the War Room added prestige to the project because this approach was novel: Moving into a conference room had been allowed only once before. Beyond that, the War Board and its assiduous updaters of the highly visible deadline chart covering its back wall recalled the drama of Battle of Britain movies with the operations room workers chalking up RAF victories and losses. Such a performance amidst an efficient but not-always-exciti ng workplace seemed to energize project group members and rally outsiders to the cause.


Nancy and Rick also used architecture to coordinate the project. The new arrangement centralized information, avoiding "multiple desks and multiple places to keep information," as Rick noted. Rick said that previously it was awkward and inefficient for him to have to rummage through Nancy's personal desk to find papers.

Also, as Nancy said, "being in a place where other people could see us helped us to grab people when we needed to when they were passing by. Helped people to stop in when they had an opportunity. . . . A lot of people have real busy schedules." An appropriately used open office places in the center the group with the most communication with others. In the next outer circle are the groups with whom the center group has the most contact. Around that are situated the groups with whom they have the most contact, etc. The point is to have the most active lines of communication be the shortest in physical length (Pile, 1978, p. 24).

Group members also had a vested interest in stopping by during one of their frequent trips past the War Room. The member's annual job review depended significantly upon the group's evaluation of her or his contribution to the project. It was very important to be "in the loop" of the group, and being close by helped.

Core Group Coalescence

A dedicated room made the project more tangible than the previous ephemeral e-mail messages and meetings in anonymous conference rooms. Duin has reported the failure of a group trying remote collaboration (1996). "Virtual teams" that "congregate" in technological space may be dehumanizing, as Raymond and Cunliffe argue (1997, pp. 14-15). Piaget notes that certain things are associated with certain places (Norberg-Schulz, p. 29). The room, to quote architectural scholar Christian Norberg-Schulz paraphrasing German philosopher Martin Heidegger, "brings something into presence" in our perception (Heidegger, 1971, p. 62, cited in Norberg-Schulz, p. 39).

Bringing the project more fully into being also brought it more fully into action. As the German philosopher Otto Bollnow noted in Mensch und Raum, "the concept of space is connected with human actions" (1963, p. 33; cited by Norberg-Schulz, p. 29). This idea is clearly expressed in the term "take place." "Place" hence has two meanings: place of action, and point of departure. It represents what is known and allows us to depart toward a "more distant goal." "Only when the individual possesses such a point (or system of points) of reference, may he act in a meaningful way," as Norberg-Schulz notes (p. 30).

Appropriating the War Room not only created a stronger sense of presence of the project to the coordinators and outsiders, but the temporary dwelling also created a "we," a subculture. Though the core group had met before the

establishment of the War Room, it was in an anonymous conference room, and though they had corresponded, it was in the immensity of cyberspace. Moreover, the SLA project had been situated "in the back forty" of a large open office. But "[human beings] cannot feel "at home" in a space without limits (Norberg-Schulz, p. 34), an important drawback because the human being is "one who dwells," as St. Exupery (1948) notes. The temporary group needed a non-virtual, human place to cohere and develop their vision of the vast project.

The site provided not only a group vision of the project but also developed the group self-image. Necessary for the development of a feeling of "weness" was a site that the group identified with. As Nancy said, the SLA's are "something we all need to feel a part of." (TR874) People need the identification provided by an "imageable structure" as Norberg-Schulz notes (p. 37). The War Room provided not only an identifiable structure but also fostered an energizing context dedicated to the SLA. Buildings, as Heidegger notes, "gather a world" and allow for dwelling (cited in Norberg-Schulz, 1986, p. 44). One must be careful, however, not to overstate this need, a distortion that can easily lead to acrimonious turf battles and thus may be akin to the Nazis' desire for "living space" or other expansionist tenets that Heidegger's philosophy in some ways supported (Harvey, 1989). The essence of the city or group is interaction, not place (Webber, cited by Norberg-Schulz, p. 27). In the case of the SLA, though, the te rritoriality was intentionally temporary. The "War Room" was operative only as long as the project. But while it lasted, the War Room gathered a cultural world including not only the project leaders and the core group with their SLA drafts and charts, but also ultimately the Vice President and other volunteer collaborators. It allowed people to dwell in the concept and goals of the SLA as they disseminated this culture in departmental collaborations throughout the company and reified this culture into the SLA documents that would guide corporate practice.

The War Board

Dominating the glassed-in War Room was the War Board, a large, highly visible chart and checklist that arrayed in its rows the 22 departments and in its 13 columns the different phases and deadlines in the collaborative writing project. As each of the 22 business unit groups completed each phase, it was checked off. The War Board was obvious to everyone who passed by, including the Vice President.

The War Board promoted the project by many of the architectural functions I have just discussed. In helping bring the project to a close, the War Board provided coordination, coercion, competition, and affiliation. The board organized both leaders and other members. Presenting her rationale for the War Board, Nancy noted, "logistically this [project] is becoming very hard for us to get our arms around. . . . We want to be sure we don't leave something out" (TR281) Pointing at the board, Rick said that it was "almost like a map. . . . This is my destination. Th[ese are] all the stops I need to make along the way" (TR994). The group now had a visible system of interrelation.

The War Board coerced by causing those who were behind schedule to suffer some trepidation in the downsizing environment. Its prominent public display in a corridor of power in MTS marshaled the symbolic capital of the building/fortress and prime internal real estate behind the SLA project. "Without having to harangue people," Nancy noted, "it's just up there for everyone to see. If you don't do it, everybody knows it" (TR 1019). Having one's performance displayed along with others for easy comparison encouraged competition of a sort. As Rick noted, "You'd be surprised how many people we had coming into the room and saying, 'Am I the last one to get my check[mark]?' . . . So there was somewhat of a competition there. At least not to be last" (TR100l).

To chart the process for everyone to see is to take the public performance of writing a step forward. Beyond writing in the fishbowl, the metaphor that Rivers (1989) used to explain organizational writing, this was closer to writing in the professional ballpark. We frequently don't keep score when we're playing for pure fun or warming up. But when we're serious and want to draw spectators, we keep score. The scoreboard created the visibility to management needed to complete the project.

Yet along with the competition with others came the solidarity of having the same mental model of the task. Also, as Klimoski and Mohammed (1994) note, "cohesion may be a consequence of team mental models," bringing forth "high effort, coordinated actions, spontaneity, assertiveness, risk-taking, etc." (p. 425). They assert that shared models increase group performance. And the SLA core group's performance certainly improved.


Architecture was an influential factor in the SLA writing process and eventually the written product because it emphasized the initial nonimportance and later strong importance of the project. Nancy and Rick wisely exploited the power of this factor to help them complete the large project under a tight deadline. Logistically speaking, architecture helped Nancy and Rick coordinate a large project by focusing their efforts on the SLA, keeping the SLA in the workflow of the organization, and presenting a visible system of interrelation for the group. The group leaders used architecture to promote group involvement, unity, and heedfulness, thus to foster a collective mind within the group. Beyond that, from the standpoint of group dynamics, "renting" a "room of its own" helped bring into presence a subculture that radiated throughout the company.

This study suggests that for a collective mind to form in large scale collaboration, we need to attend to (a) the form of the mental representation of the text and (b) the architectural forms that provide a space for the collective mind to coalesce. But how may we address these concerns in our classrooms and on the job?

Implications for Pedagogy and Practice

Robert McEachern (1997) notes that several studies of bad news letter writing (including his own) suggest that textual formulas are "not effective in and of themselves but as models to be adapted according to the rhetorical problem presented by the situation" (abstract, p. iv). What my study suggests is that after the group has a preliminary fix on the rhetorical situation of its project, it is better for someone to generate or appropriate some kind of text to use as a point of reference for any group writing. Because of their tangibility, texts seem to do more than do oral concepts to help a group ultimately "get" the form and perceive and accommodate a schema that can then assimilate additional information and strategies in successive drafts.

This study also suggests that we need to teach students and practitioners not only how to carefully read texts, but also how to read buildings closely and appropriate those spaces to catalyze writing groups

Implications for Further Research

We should test the effectiveness of having groups of students, practitioners, and perhaps even individuals, do close imitations of business model texts as a prerequisite for learning forms and for writing the same kind of document situated in a real organization with a real problem. The process would be (a) imitate model text closely or outline it, (b) abstract rules from the text, (c) apply rules to new context (local organization's problem). Groups may need to learn to imitate situated forms, not just general patterns of organization, before they can transform forms to meet the exigencies of another context.

Finally, more research needs to be done to determine, first, the extent of large-group collaboration in industry; second, what kinds of large groups exist; and third, the strategies that, for a large writing group, form a successful collective mind.

Geoffrey A. Cross received the NOTE 1995 Best Book in Scientific and Technical Writing Award for his book Collaboration and Conflict, and the 1997 ABC Outstanding Researcher Award. Earlier portions of this article were delivered at the 1996 and 1997 international ABC conventions; a preliminary version of the entire paper was presented as the Outstanding Researcher Lecture at the 1998 ABC international convention. This article is excerpted from his new book, Forming the Collective Mind (Cresskill, NY: Hampton Press).

This research was funded by grants from the Philip Rauch Center for Business Communications, Lehigh University; the C. R. Anderson Research Fund of the Association for Business Communication (ABC), and the University of Louisville. The researcher wishes to express his gratitude to all organizations for their support. The researcher also thanks Anthony Baker for his precision and initiative and Vicky Cummings for her steadfastness in the coding and data entry of drafts. Thanks also go to Kitty Locker and the anonymous reviewers of my article and book manuscripts for their helpful comments.


(1.) While writing the history of Section A, I sorted the changes several ways to identify important patterns. Looking, for example, at each change made in each document and identifying its kind (e.g., insertion) helped me trace the SLA evolution from outside document and very specifically what role each draft of Section A played in that progression.

For example, statistics regarding Draft 6 suggested that it was a major conceptual change from its predecessor. Looking at the numbers was not enough, however; they had to be followed by an analysis of the substance of the changes in the context of the evolving document and situation. The sort statistics told me that more insertions occurred in Draft 6. But reading the document after the meeting with CIF showed that although the document now described CIF, that description suited MTS' needs to learn about CIF rather than CIF's need for understanding MTS' rules of operation-these were contained in another section of the document. In sum, the data entry, sorting, and finding patterns were necessary but not sufficient for accurate reporting and theorizing about textual progression in large-scale collaboration.

This fine-grained approach was helpful in many tangible ways and also served to help me "get close to the data" for more holistic understandings. I used the material to write this history of the drafting, revision, and editing of the Service Level Agreement.

All subjects have been given pseudonyms. An agreement has been signed that this study will not be used to interfere with the career of any human subject in the research.


Asch, S. (1952). Social psychology. Englewood Cliffs, NJ: Prentice-Hall.

Becker, F. D. (1981). Workplace: Creating environments in organizations. New York: Praeger.

Bourdieu, P. (1991). Language and symbolic power. Cambridge: Harvard University Press.

Brewer, M. B. (1988). A dual process model of impression formation. In T. K. Srull & R. S. Wyer (Eds.), Advances in social cognition (Vol. 1, pp. 1-36). Hillsdale, NJ: Lawrence Erlbaum.

Broadhead, G. J., & Freed, R. C. (1986). The variables of composition. Urbana: NCTE.

Cross, G. A. (1994). Collaboration and conflict Cresskill, NJ: Hampton Press.

Doheny-Farina, S. (1984). Writing in an emergent business organization: An ethnographic study. Dissertation Abstracts International, 45, 3337a (University Microfilms No. 85-00,944)

Doheny-Farina, S., & Odell, L. (1985). Ethnographic research on writing: Assumptions and methodology. In L. Odell & D. Goswami (Eds.), Writing in nonacademic settings (pp. 503-525). New York: Guilford.

Duin, A. H. (1996, November 6-9). Collaboration via the internet: Cross-cultural student teams from Australia, Norway, and the U.S. Paper presented at the international convention of the Association of Business Communication, Chicago, IL.

Ede, L., & Lunsford, A. (1990). Singular texts/Plural authors. Carbondale: Southern Illinois University Press.

Forman, J. (1991). Collaborative business writing: A Burkean perspective for future research. The Journal of Business Communication, 28, 233-257.

tici Giilmore, T. N., & Barnett, C. (1992). Designing the social architecture of parpation in large groups to effect organizational change. The Journal of Applied Behavioral Science, 28, 534-548.

Hammer, M. (1996). Beyond reengineering. New York: HarperCollins Business.

Harvey, D. (1989). The condition of postmodernity: An enquiry into the origins of cultural change. Cambridge, MA: Blackwell.

Hildebrandt, H. (1995, November 1-4). Business communication and architecture: Is there a connection? Paper presented at the Association for Business Communication International Convention, Orlando, FL.

Kleimann, S. D. (1989). Vertical collaboration and the report review process at the U.S. General Accounting Office. Dissertation Abstracts International, Vol. 51-03A, p. 0833 (University Microfilms No. 9021588).

Klimoski, R., & Mohammed, S. (1994). Team mental model: Construct or metaphor? Journal of Management, 20, 403-37.

McEachern, R. W. (1997). Writing "bad news" in an arts organization: A socio-cognitive inquiry. Dissertation Abstracts International, Vol. 58-07A, p. 2632 (University Microfilms No. AAG9801885).

McIsaac, C. M., & Aschauer, M. A. (1990). Proposal writing at Atherton Jordan, Inc. Management Communication Quarterly, 3, 527-560.

Miles, M., & Huberman, A. M. (1994). Qualitative data analysis. Beverly Hills: Sage.

Newell, A. (1990). Unified theories of cognition. Cambridge, MA: Harvard University Press.

Norberg-Schultz, C. (1986). Architecture: Meaning and place. Milan: Electra Spa.

Piaget, J. (1971). Biology and knowledge (B. Walsh, Trans.). Chicago: University of Chicago Press.

Piaget, J. (1981). Problems of equilibration. In J. Gallagher & D. K. Reid (Eds.), The learning theory of Piaget and Inhelder (pp. 210-220). Austin: Pro-Ed.

Pile, J. (1978). Open office planning. New York: Whitney Library of Design/ Watson-Guptill.

Raymond, S., & Cunliffe, R. (1997). Tomorrow's office. London: E & FN Spon.

Rivers, W. E. (1989). From the garret to the fishbowl: Thoughts on the transition from literary to technical writing. In C. B. Matalene (Ed.), Worlds of writing (pp. 64-80). New York: Random.

Smith, J. B. (1994). Collective intelligence in computer-based collaboration. Hillsdale, NJ: Lawrence Erlbaum.

St. Exupery, A. (1950). The wisdom of the sands. (S. Gilbert, Trans. from the French Citadelle.) New York: Harcourt Brace.

Tebeaux, E. (1996). Nonacademic writing into the 21st century: Achieving and sustaining relevance in research and curricula. In A. H. Duin & C. J. Hansen (Eds.), Nonacademic writing: Social theory and technology (pp. 1-16). Mahwah, NJ: Lawrence Erlbaum.

Weick, K. E., & Roberts, K. (1993). Collective mind in organizations: Heedful interrelating on flight decks. Administrative Science Quarterly, 38, 357-381.

SECTION I-Description of Business Unit requirements

Department XYZ is in the business of providing Life Insurance Products to customers residing in Southeastern United States. This department utilizes a number of Mainframe Systems to process an estimated 50,000 transactions a month. To facilitate timely processing and meet policy holder needs the business unit requires:

On-line Systems be available from 7:00 am to 6:00pm Monday thru Friday

Hardcopy reports be generated and delivered by 9:00 am Monday thru Friday

Command Center Support on Priority 1 calls be resolved within one hour

Support Hotline Support on Priority 1 calls be resolved within two hours
Agreements / Expectations Reports & Information
Workstation Service       Customer subscribes to this Service as
                          outlined in the Standard SLA.
Desktop Services          Customer subscribes to this service as
                          outlined in the Standard SLA.
Network Services          Customer subscribes to this Service as outlined
                          in the Standard SLA, with the exception of LAN
                          Support which will be provided by their business
Software Administration   Customer subscribes to this Service as
Services                  outlined in the Standard SLA.
Mainframe Services        Customer subscribes to this Service as
                          outlined in the Standard SLA.
Warrantdat                Needs Claims Recon Report by 6:00 a.m.
KOL                       Needs Checks by 7:30 a.m.
HDK                       Needs to be able to have 4 hour tirm-around on
                          "Prioity Requests"
(Same format for Consulting and Performance)
MTS & IS are responsible for ensuring the environment and integrity of the
     following Systems are maintained in a highly satisfactory manner.
System(s) [*]:
VIR System          Warrantdat    Keywords Search (Omega)
Range Manageability Payrollee/OCR ZKD
Cemter              Concav
(*.)Each system is documented in Section II.

Customer Responsibility

The inability to deliver a level of service equal to or above the stated target, which results in a negative business impact as defined by the customer, constitutes an exception to the performance standard for the month. All exceptions will be logged, researched, and detailed in the monthly measurement reports.

TS-need. All service disruptions will be logged in a Lotus Notes data base. Customers who have access to Lotus Notes are encouraged to log disruptions and their impact as soon as the problem is resolved. Customers who do not have access to Lotus Notes may call the command center and have their disruption logged.

Note: The service disruption logging procedure is NOT a replacement for the command center. Problems must still be reported through the normal problem resolution process. Any disruption which has caused a measurable, negative business impact will also be entered into the service disruption data base if the customer requests the issue to be logged. The service disruption logging procedure is designed to report against performance objectives and trend analysis.

Responsible for ensuring all current day transaction entries are entered by 5:50pm.

Responsible for verifying that all outputs are correct and received on schedule.


Exception Reporting. Any party (MTS, IS, or Client) is responsible for timely reporting of any exception to the agreements outlined in this document. Exception calls should be reported to the Command Center (Mainframe) and the Support Hotline (PC LAN Environment). MTS and IS groups will utilize a Contact List to notify the appropriate Clients. Clients will need to notify individuals in their respective work areas.

MTSIIS Meetings. A morning status meeting will be held daily Monday through Friday (8:30 a.m.) in the Magnolia Room of the Data Center building to discuss problems experienced with the previous nights processing. IS Representatives should be in attendance when ever their system encounters problems.

A weekly change meeting will be held on Thursday (10:00 a.m.) in the Magnolia Room of the Data Center building to discuss changes to the production environment. MTS and the IS groups will discuss the tentative calendar of events, implications and Customer impact. Go/No Go Decisions will be made in this meeting. Information from this meeting will be documented and posted on bulletin boards by noon.

Monthly Reporting. A monthly service level report will be produced. The report will include the defined service level objectives versus the actual and a comment section to explain variances. The report will be available by the 12th business day of the month. (NOTE: SURVEY RESULTS INDICATED THAT CUSTOMERS MAY ONLY WANT DATA BI-MONTHLY)

Annual Review. MTS will schedule an annual review of SLA's with each Business Unit Contact Person. This document will be reviewed prior to expiration and or when major changes occur to the system(s) identified. Parties to this agreement can review this document at any time. A major change is one that will have a severe impact on availability, reliability, and/or response time.

Re-Negotiation Procedure. Re negotiations should normally commence two months prior to any major changes in business requirements. MTS will schedule discussions with the Client group and IS group(s). In the case of "low impact" changes this agreement may be amended. Amendments are still negotiated with the Client, IS, and MTS. They are signed, dated, and attached to the end of this agreement. They should be titled "Amendment A" and continue with "B". and "C". etc. as needed.

Consolidated Insurance Financial

Consolidated Financial associates in Laketown and Metroville provide support for Actuarial, Audit, Employee Services, Financial Reporting, Financial Services, Planning, and Reengineering. Consolidated Insurance is organized into three legal entities: Heric and Blain Insurance, Aix en Provence Insurance, and Pornic Insurance. Financial work is very technology dependent. Network and system availability is critical to the daily management of financial data.

Mission Statement

Consolidated Insurance's business is providing insurance and related services to middle and lower income consumers based on face to face contact.


Consolidated Insurance Actuarial use a variety of mainframe tools to manipulate and report on data generated by financial insurance processing, and field accounting systems. Information modeling, statistical programs, and projections are the basis for analysis and decision making. The heaviest actuarial workload occurs at year end and quarter end during financial reporting periods. In addition, projects requiring intensive analysis and processing are often initiated on short notice. Mainframe availability, access to data and response time are critical factors in performing actuarial work.


Internal audit evaluates the extent and effectiveness of management controls, both manual and automated. Associates in audit work with project teams on system conversions and provide consultation to management for proposed manual and system changes. Other activities include extracting data, running reports, developing software, and performing testing. Mainframe availability to run risk models at quarter and year end is critical. Additional DASD is required at month end and year end.

Financial Reporting

Year end and quarter end are peak times for financial reporting. LAN, MAN, and WAN availability are critical at these times. Mainframe availability is critical and response time is very important for completing transactions in the hourly upload "sweeps".
COPYRIGHT 2000 Association for Business Communication
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Cross, Geoffrey A.
Publication:The Journal of Business Communication
Article Type:Brief Article
Date:Jan 1, 2000
Previous Article:Work Values, Cognitive Strategies, and Applicant Reactions in a Structured Pre-Employment Interview for Ethical Integrity.
Next Article:Chinese Business Negotiating Style.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters