Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Engagement with text : collaborative writing in a high technology company Begoray, Deborah Leslie 1994

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1994-95311x.pdf [ 3.54MB ]
JSON: 831-1.0054892.json
JSON-LD: 831-1.0054892-ld.json
RDF/XML (Pretty): 831-1.0054892-rdf.xml
RDF/JSON: 831-1.0054892-rdf.json
Turtle: 831-1.0054892-turtle.txt
N-Triples: 831-1.0054892-rdf-ntriples.txt
Original Record: 831-1.0054892-source.json
Full Text

Full Text

ENGAGEMENT WITH TEXT:COLLABORATIVE WRITING IN A HIGH TECHNOLOGY COMPANYbyDEBORAH LESLIE BEGORAYB.A. (Hon) The University of Alberta, 1975M.A. The University of calgary, 1985A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIES(Centre for the Study of Curriculum and Instruction)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAAugust 1994© Deborah Leslie Begoray, 1994In presenting this thesis in partial fulfillment of therequirements for an advanced degree at the University of BritishColumbia, I agree that the Library shall make it freely availablefor reference and study. I further agree that permission forextensive copying of this thesis for scholarly purposes may begranted by the head of my department or by his or herrepresentatives. It is understood that copying or publication ofthis thesis for financial gain shall not be allowed without mywritten permission.(Signature)Department of ( L7rL 1 jç ,qL( rr7Cu .Z’tru cII>The University of British ColumbiaVancouver, CanadaDate J/Q’/•1”iiAbstractOver the past decade, an interest in collaboration hasbeen coming to the fore in composition studies. Whereas oncewe were primarily interested in investigating the cognitiveprocesses of the individual, we now seek to understand moreabout the social dynamics of writing in groups to improve ourteaching of composition in the classroom. To that end, thisdissertation looks at the real world collaborative activitiesof business proposal writers within a high technology company.Writing in the workplace is often undertaken in groups, and mywork at Cerebellum, Inc. with computer professionals (whowrote as part of their jobs) reveals complexities hithertounsuspected in the social writing process.The importance of a detailed understanding ofcollaboration has been called for in the literature by, forexample, Ede and Lunsford (1990). My dissertation surveyscurrent literature in composition, including a review ofinvestigations into collaboration during business writing as asalient behaviour of such a discourse community. In order toaccomplish my research, I used a video camera to record theactivities which embodied the writing process at CerebellumInc. I found that the use of the video camera in anethnographic manner not only helped me to gather detaileddata, both verbal and nonverbal, in the continuous andcomprehensive detail so vital to communication research, butalso assisted in initiating better understanding within theiiibusiness community of the aims and approaches of academicresearch. Video technology gave me a chance to participate inas well as observe situations, and also opened the door toconversation concerning my methods and my findings with bothresearchers and informants.I propose a model of the varying levels of engagementundertaken by the writers of a business proposal. I thensuggest the educational value of the representation with adiscussion of implications for the teaching of writing in theworkplace and in more traditional school settings.Detailed research into collaboration offers us a windowon the social processes which constitute writing for ourstudents now and in their futures in the workplace. Such workis vitally important to ensuring superior levels of advancedliteracy which will be in continuing demand now and in thenext century.ivTABLE OF CONTENTSAbstract iiTable of Contents ivList of Tables viiList of Figures viiiAcknowledgements ixChapter One Introduction and Overview 1Studying Writing as a Social Process 5Understanding Collaboration in theWorkplace 8Organizational Communication 9Research Questions 11The Research Context 13Proposal Writing at Cerebellum, Inc. 14Education, Research and Society 15Chapter Two Review of Professional Literature 21Cognitive Research 22Sociocultural Research 28Audiences and Communities 29Organizational Communication 40Collaborative Writing 43Specific Research Findings 46Questions Remaining 50Chapter Three Methodology 54Gaining Access to Data 57Accessing the Research Site 58Accessing the Lives of Informants 60Gathering Data 61Sample Selection 68Sorting Data 70Analyzing Data 72Representing Data 76Chapter Four A Model of Levels of Engagement withText 80The Business Proposal 80Levels of Engagement with Text 81Level IlI--Strategizing 85Ongoing Themes 87VLevel 11--Information Gathering 90Ongoing Themes 92Level I--Drafting 95Ongoing Themes 96Conclusion 98Chapter Five Presentation and Discussion of EarlyEvents 101The Borden Proposal 102Dramatis Personae 102General Background 103The Sundial Proposal 104Drarnatis Personae 104General Background 105Day One 107Event A The Starting Gun 108Event B Opening Moves 122Days Two and Three 128Day Four 129Event C Changing Horses 129Event D Who’s on first? 131Event E Decisive Action 133Day Five 134Event F Under Consideration 134Event G Quick Change Artistry 141Chapter Six Presentation and Discussion of MiddleEvents 142Day Six 144Event H Problems, Problems 144Event J Down to Work 163Day Seven 165Event K Covering Of f andEvent L The Audience Awaits 165Event M A Decisive Dyad 171Day Eight 174Event N Collabo-tech 174Event P Enter the Dragons, Part A 175Chapter Seven Presentation and Discussion of LateEvents 182Event P Enter the Dragons, Part B 182viDay Nine 187Event Q The Recyclers 187Event R A Woman’s Place 189Event S Pulling Rabbits from Hats 193Day Ten 195Event T Asking the Tough Questions 196Event U Rounding Up the Stragglers 205Event V Words, Words, Words 207Event W If Only 209Event Y Coming Together 210Day Eleven 212Event Z End Game 212A Postscript 215Chapter Eight Conclusions 217Level III Findings 219Level II Findings 231Level I Findings 244Chapter Nine Implications 260General Implications for AdvancedLiteracy Development 261Implications for Reading, Writingand Using Texts 263Implications for Representing Audienceand Community 267Implications for Further Research 270Bibliography 274viiList of TablesTable 1. Outline of Levels of Engagement 83Table 2. Classification of Events by Level, Day,and Participant (in order of rank) 100viiiList of FiguresFigure 1. Path of Document Movement Between Levels 99ixAcknowledgementsThis research, of course, was possible only with thegenerous cooperation of “Cerebellum Inc.” and the employeeswho wrote the “Borden” and “Sundial” proposals.I owe a debt of gratitude to the members of my committee(John Willinsky, Judy Segal, and Ricki Goldman-Segall) fortheir valuable guidance during the preparation of thisdissertation.My thanks always to my husband, John Begoray, forinsightful comments, continuing support and pep talks.To those who seek more evidence that writing is acollaborative act, I offer these acknowledgements.In addition, I wish to acknowledge the financial supportof a Doctoral Fellowship through the Social Sciences andHumanities Research Council of Canada.1CKAPTER 1: INTRODUCTION AND OVERVIEWIn the business world, general communication abilities,both oral and written, have always been valued. They areacknowledged as useful and necessary foundations for thecommonplace activities of professional life in corporateCanada: making presentations, participating in meetings,negotiating business deals, writing letters and readingpromotional materials.As businesses have become more specialized, so too havethe demands on the literacy skills of employees. While thetechnical personnel (such as technicians and engineers)employed by computer companies, for example, are paid to solveproblems, the scope of their job has changed. Where once talkand reading may have constituted the extent of the languagedemands on the technical staff, many of these employees nowwrite as part of their job (Faigley and Miller, 1982), and arerequired to do so in more and more sophisticated ways.With thousands of new products flooding the computermarketplace, demand is high for technicians with in-depthknowledge about a number of products, who can continue to addto their knowledge on the job by reading trade journals andtexts. Concomitantly, the industry also needs technicalwriters, and technicians who are writers. These employees maybe called upon to compose manuals for users of the plethora ofnew products being developed at many levels of expertise:beginners to sophisticates. Effective documentation has taken2its place as a key differentiator in the competition amongcomputer products.The nature of the workplace in the nineties has alsochanged in other ways for those working in technologicalareas, as computer technicians do. Not only must they beeffective technically, and be able readers and composers oftechnical documentation for a variety of products, but theymust also understand marketing. Mclsaac and Aschauer (1990)observe that engineers, for example, strongly prefer writing“where technical problems reign supreme, where conclusions areself-evident, where deliberate persuasion is suspect” (p.550) . Such attitudes, which show technical writing asdivorced from presentation of the business or sales case, arefading in the bright light of new mandates for achieving acompetitive edge. Technicians who write about their own worksoon discover that the “crafting of language for promotionalpurposes in the technical environment is everyday fare nowjust as it has been throughout the history of consumermarketing” (Bryan, 1992, p. 76).Indeed, the new cry in the business world is for “totalquality” in all aspects of product and service offerings.Driven by the work of theorists such as W. Edwards Deming(1988), many business leaders are developing in-housestandards which will allow them to be recognized as operatingwithin total quality controls. The time is rapidlyapproaching when businesses must qualify for certificatesgranted by external agencies such as the International3Standards Organization (ISO 9000, 1991), guaranteeing theiradherence to quality standards, including superior writing, inorder to do business with many government agencies and largecorporations.Japanese approaches to quality control are hastening thisdrive to standardization. Despite more recent volatility inthe Japanese economy, business practices long admired bywestern companies for their success in Japanese enterprisesare now being adopted in North America. Total Quality Control(TQC), a blend of eastern and western visions of superiorbusiness practice, is now a goal for many businesses.Communication, according to the principles of TQC, must be“decentralized, vertical-upward, interdepartmental,interdependent, trusting, long-term, group-oriented,reciprocal, immediate, nurturing feedback, flexible, andcharacterized by close proxemics” (Goldman, 1993, p. 29). ATQC approach involves the breakdown of traditional barriersamong the many and varied communicators who work in business.They must not only develop skills in reading each other’swork, but also assist in the development of documentsinvolving a variety of writers, the vast majority of whomwrite as a condition of their job (rather than asprofessionals dedicated to writing).Communication in modern business involves not onlyemployees but also, of course, clients. No longer is itsensible to make a sale with attention paid only to the bottomline profit. Now what is valuable are the employees who4contribute to the presentation which sells the present job andalso builds a foundation of good faith in hopes of futuresales. Technicians, sales personnel, and managers have beenforced to recognize that in a highly competitive economy,still suffering as Canada is from the effects of recession,“on-going success in sales requires building relationshipswith clients” (LaDuc, 1991, p. 157) . One way suchrelationships are achieved is through ability in writing thosekey documents, such as business proposals, which connect onebusiness with another. For such documents, where the demandfor quality is high, research shows that companies often usecollaborative writing approaches (Couture and Rymer, 1989).In an attempt to further elucidate the compositionalpractices of those who are working together on a businessproposal in a high technology business community, I have•collected and analyzed data which suggest that writers workingtogether experience different levels of engagement with theemerging proposal text. These levels (strategizing,information gathering and drafting) I have represented in amodel. A schematized version of the social writing processwill provide a concise and practical beginning point for adiscussion of ways. to better carry out (and help others tocarry out) writing as a collaborative activity in a variety ofsettings.5Studying Writing as a Social ProcessAn important movement in writing research involves a refocusing of attention away from the individual writer inisolation (for example, Flower and Hayes, 1981) and towardsthe social context as a factor influencing the composition act(Bizzell, 1982; Bruffee, 1986; LeFevre, 1987) . Just howcrucial the social role is differs from group to group ofcomposition researchers; nevertheless, the trend in theprofessional literature is clear: “no man [or woman, either]is an island/entire of itself”, and should not be studied asthough they were.This is not to say that the only important gains made inour understanding of writing have come in recent times, withcurrent trends, and with a growing understanding of, or atleast more attention to, social processes. Our interest indiscovering how communication might be shaped dates from atleast the time of Aristotle’s Rhetoric (1932), itself a resultof the influence of Plato and the Sophists on thecommunication habits of Greek society in the second century.Consideration Qf writing, however, has largely focused on theproduct and what it might contain in arguments (logos), inorder to appeal to a specific audience (pathos), and bedelivered by a rhetor skilled in writing or speech-making(ethos). Certainly, creating the document required skill insuch matters as choosing ideas, and organizing themeffectively, but the focus remained on the ultimate impact ofthe completed product.6In the late twentieth century, the concerns of writingresearchers shifted. In 1971, Janet Emig published TheComposing Processes of Twelfth Graders, an event sincedescribed (see for example, Purves, 1992) as heralding achange in the primary focus of composition theory and researchfrom an interest in written products to an interest in thewriter’s composing processes. Emig borrowed much of herresearch methodology from the field of cognitive psychology,such as case study techniques, to discover what the individualwriter did as composition proceeded. By 1981, Flower, arhetorician, and Hayes, a psychologist, published a cognitiveprocessing model of composition. The resultant outpouring ofresearch activity based on this model established thecognitive psychologists as a central force in the study ofwriting.Not all writing researchers were satisfied with thecognitive psychological approach to studying the writingprocess, however. One of the difficulties raised withinmonths of the model’s publication was its seeming lack ofconcern with the influence of the outer-world upon thecommunicative act: “What’s missing here is the connection tosocial context afforded by recognition of the dialecticalrelationship between thought and language” (Bizzell, 1982, p.223). If, indeed, language is modified by our interactionswith others, which in turn influence our thoughts, whichfurther modify our communicative acts, then social effects on7writing are certainly important (Halliday, 1969; Vygotsky,1978)Both research and scholarship on the writing process nowreflect a concern for social contexts from a broad variety ofresearchers. Bizzell and others, especially Bruffee (1983,1984, 1986), have evolved a theory of social construction toexplain the writing process. The origins of this school ofthought, however, come far earlier. Kuhn’s The Structure ofScientific Revolutions (1970) postulated that scientists, as acommunity, are the “producers and validators of scientificknowledge” (p. 178). When enough members of the group changetheir minds regarding an idea, a paradigm shift occurs. Theinsights of philosophers such as Rorty (1979) and culturalanthropologists such as Geertz (1983) heightened the interestof composition scholars in examining the influence of thegroup upon the writing act, and the writing act upon thegroup’s beliefs.Cognitive psychologists have more recently found greatmerit in adding a concern for context to their work withindividual writers (Langer, 1985; Freedman, Dyson, Flower,Chafe, 1987; Flower, 1989), dubbing their researchsociocognitive. The name, it seems to me, is significant: theinfluence of the social upon the cognitive. Socialconstructionists concentrate more upon the group influenceresulting in a document written by one or more group memberswhich reflects the group’s beliefs. Rhetoricians of thissocial constructionist group (unlike Flower, a8sociocognitivist) study issues of the functioning of thediscourse community (Bizzell, 1987; Harris, 1989) and itsconstruction of knowledge: the social epistemic (Lunsford,1992) . They, too, however, have acknowledged that theindividual community member cannot and should not beforgotten. FrQm such individuals comes the nonstandard, orabnormal discourse (Rorty, 1979) which keeps the group frommerely reproducing itself. Through stimulus and conflict(Trimbur, 1989; Lyon, 1992) among members, groups reach theirbest performance.Understanding Collaboration in the WorkplaceThe trend to inquiry in the social processes of thecomposition is especially evident in the work of researcherswho investigate the professional writing community. Aspreviously discussed, Kuhn’s (1970) work aroused interest inthe writing community and quite naturally lead toinvestigations of the writing processes of such professionalsas scientists (Bazerman, 1988) , bankers (Smart, 1993), andbusiness professionals (Faigley and Miller, 1982; Odell andGoswami, 1985; Doheny-Farina, 1986; Couture and Rymer, 1989;Cross, 1990)If writing is indeed a social process, then inevitably,perhaps, research attention would focus on writers workingtogether. Ede and Lunsford’s (1990) monumental project, whichlooks at writers collaborating in seven professionalorganizations, is but a recent addition to research in9collaborative efforts in composition. Research by survey andinterview has given us ample evidence that collaboration inthe workplace is frequent and widespread (Faigley and Miller,1982; Anderson, 1985; Ede and Lunsford, 1990). Ethnographicexamination is beginning to take a more in-depth look at thenuances of the social processes of writing in specificsituations (Doheny-Farina, 1986; Cross, 1990), with much workstill needed.Organizational ConimunicationCollaborative writing is a complex activity whichrequires an examination of ideas in several associated areas.Superior communication within companies leads to betterdocuments. The effective business proposal must often becomposed by a team of writers. Thus, business writers must beable to function well within a group (Doheny-Farina, 1986;Cross, 1990; Stohi and Schell, 1991), or even within severaldifferent groups working on various projects simultaneously.Each group, of course, presents its own set of challenges inachieving the desired characteristics of interdependence,trust and flexibility which will help lead to effectiveconstruction of a winning product. Writing together is one ofthe most complex tasks which individuals can face, involvingmany activities which must lead ultimately to clear, accurateand convincing documents.10Business proposals must present to their audience aunified voice. Consensus during the writing process, however,while laudable and certainly a goal for any group project,must be tempered by ideas arising from the conflict of varyingviewpoints encompassed by the group. Members must, therefore,learn to maximize their individual contributions to the team,despite pressures to give in too quickly to “groupthink.”Various factors, some researchers speculate, such as thepresence of a highly vocal, talented and “titled” employee canlead to other group members minimizing their input (Veiga,1991) . Studying writing in the workplace must attend to suchproductive and nonproductive relations among group members.What the field of collaborative writing research seems torequire most, however, to help examine group writingapproaches is progress toward the building of a model ofcollaborative writing. Nystrand (1989), in proposing asocial-interactive model of writing, suggests that “Ourinterests in understanding writing and written communicationrequire that we bring order to complexity by elucidating basicprinciples and regularities in a relatively parsimonious andsimple form” (p. 81) . Such interests are not to deny thecomplexities of the writing processes which surroundcollaborative writing, but rather to begin to look forpatterns in the process. Nystrand is concerned mainly withthe version of “truth” which arises through the interaction ofwriter and reader as mediated by the text. Obviously, modelsof social processes must also be constructed to focus our11examination of document composition by several writers workingtogether.One place to begin looking for patterns from which tobuild a representation of the collaborative process is in theclose analysis of events surrounding the production of asingle document by several authors. Smudde (1991), in hiswork with General Motors, suggests that a writer moves fromconception to final product through several stages. Duringeach phase, the writer enlarges an understanding of theproject. However, Smudde confines his attention tocollaboration by limiting it to one possible cause of awriter’s expanded view. The writer is still presented bySmudde as an isolated inventor. What is needed now is a modelwhich includes all of the collaborators whose work defines thefinal document, and shows also their activities in concertwith one another. Perhaps in this way, we might reach ajustifiable balance between the complexities of anethnographic “thick” description (Geertz, 1973) and the overlysimple models proposed by Nystrand (1989) and Smudde (1991).Research QuestionsAn examination of the current state of research inprofessional, nonacademic writing indicates that what wouldcurrently be helpful would be an investigation further testingour assumptions about the collaborative habits of businesswriters and then representing these habits in some pragmaticand pedagogically useful form. An ethnographic approach to12data gathering would offer far greater detail to build such amodel than would the survey approaches which currentlydominate our understanding. Using a video camera to gatherextensive data of a single writing process (such as thecomposition of a business proposal) would further heighten thechances for in-depth analysis, based on repeated viewing, todiscover nuances of meaning expressed verbally and nonverballyin a communication situation. The social-construction-of-writing theory suggests that writers work together to buildmeaning. An examination of the interactions among people indifferent positions in a business should provide insightsabout just how those employees construct meaning whilecollaborating on a document.During two years of research at a high technology companywhich I have called “Cerebellum,” I hoped to achieve a greaterunderstanding of the social writing process. My specificresearch questions, designed to focus my investigation, werethe following:1. What are the collaborative manners and activities ofwriting involved in producing a business proposal• document intended by one business to persuade another?2. How might the proposal writing process, demonstrated by acollaborative writing group, be revealed, represented andimproved by a model based on ethnographic data gatheredin a typical high technology setting?13The Research ContextThe corporate setting for my investigation intocollaborative writing was “Cerebellum Inc.” Cerebellum is alarge, Canadian-based computer systems integration firm. Ithas branches in most provinces and in several other countries.As a systems integrator, Cerebellum brings together computersoftware and hardware to create high-technology solutions tobusiness problems for large private companies and governmentoperations. It works extensively with sub-contractors as wellin an effort to provide a total business solution for clients,coordinated by Cerebellum managers. These solutions areinstituted by teams composed of employees from Cerebellum,from the sub-contractors, and from the client operation.Cerebellum may also train client employees who will use thesystem, and/or maintain it (e.g. fixing it if it fails tofunction properly) once it is up and running. Projects helpto establish the long-term relationship so prized by manycompanies, often resulting in future work for the same clientwhen they upgrade or add to their existing system.The Vancouver branch office of Cerebellum has offices ontwo floors of a downtown office tower. It is a glamorousmarble and glass building with a view of both mountains andocean. At the conclusion of my study, the office had nearlyninety employees, having undergone significant growth in thetwo years during which I conducted this research. Cerebellumoften moves workers between branches or out to client sitesfor extended periods of time. As I became more intimately14acquainted with these Cerebellum writers, I developed anincreasing awareness of the role of collaboration in the lifeof a professional who writes on the job. Not only were truthsbeing explored in the e-mail messages which flowed back andforth through collaboration between writers and readers, asNystrand might.note, but the business employees’ jobs ingeneral were suffused with collaborative writing. They wrotetogether, even when they were alone.Proposal Writing at Cerebellum, Inc.Cerebellum competes regularly for work, and collaborationfigures largely in their business strategizing. The “classic”(their term), though certainly not the only, approach topersuading a client to engage their services is by respondingto a Request for Proposal (RFP) with a business proposal.These proposal documents range in length from several pages toseveral binders. The size often varies with the potentialvalue of the business. Proposal writing teams, or “task-groups,” are assembled to compose the document. These groupsvary in size, but commonly feature similar members such astechnical staff, sales personnel and managers. Each memberbrings expertise to the group, expertise without which theproposal could not be written.Final drafts are ultimately the responsibility of theaccount executive, but individuals take responsibility for thecontent (if not for the style) which they have contributed.The technical architect, for example, is responsible for15ensuring that the solution to the technical problem can beguaranteed to work as specified. Technicians contributing todocuments must write carefully to promise only what can bedelivered. Proposals thus become delicate balancing actsbetween suggesting solutions which will differentiateCerebellum from its competitors, thereby taking on some risk,and making certain not to promise answers which cannot beimplemented in order to protect Cerebellum from risk. Thus,the proposal, which focuses a group of writers on an importantcollaborative task representative of text production atCerebellum, provides a focal point (or “test case!!) for myresearch into some of the larger issues of the social writingprocess.Education, Research and SocietyAlthough the proposal provides a focus for my research,I also wish to place such a collaborative writing project in alarger context. As Canadian society struggles to re-formulateitself to face the challenges of a global economy, hampered bya large and growing deficit and a populace discouraged byunemployment, education seems an obvious place to look foranswers. Statistics Canada figures show that jobs in hightechnology areas (perhaps as many as 300,000) are presentlyunfilled for lack of qualified workers. The people who mighttake up these positions will be required to fulfil not onlythe technical demands but also the new demand for advancedliteracy. It thus becomes urgent to investigate questions16surrounding workplace communications. These inquiries mustacknowledge and address the often problematic relationshipsamong education, educational research and the needs ofsociety.First, whether research looks at academic or nonacademicvenues, investigators must develop a sensitivity to informantswithin a social context. Classroom or executive suite,playground or boardroom: people construct their own contextsand are in turn constructed by them. Research approachessuitable to the study of complex cultural behaviours become apart of that context and will have an impact on subjects andtheir milieu. Howe and Eisenhart (1990) caution allresearchers that “[tihe research process itself must giveattention to the nature of the contexts and individuals itinvestigates and to which its results might be applied, thatis, to their social, political, and cultural features” (pp. 7-8)Second, not only do these effects apply to the informantculture but also to the research community which will openitself to reciprocal gains as they interact with theworkplace. Sensitive research will acknowledge its oftenhegemonic claims to authority and modify these: “As members ofthe research community, we need to understand the way ourdisciplinary discourse appropriates the experience of theresearch subject and represents it in our institutions”(Herndl, 1991, p. 320) . Academic issues, such as literacy,find unique realization in the corporation. Appropriate17research techniques will be those which will representexperience accurately and fairly, and benefit both researchand business communities.Third, educational research, which strives in this caseto link the concerns of two often disparate and occasionallymutually suspicious groups, must also provide for educators atlarge to understand and consider the possible implications ofthe investigation for students: “the language of the resultsand implications must be in a form that is understandable to,and debatable by, various actors in a particular setting--teachers, administrators, parents, and also educationalresearchers with varying perspectives and expertise” (Howe andEisenhart, 1990, p. 7). Models arising from the deep analysisof audio-visual data are one readily accessible way to presentfindings to a wider community. Although the nature of mystudy permits no broad generalizations, I do neverthelesspresent my findings for further confirmation in hopes thatcases of a number of typical collaborations will eventuallylead to generalizability. Classrooms, whether in schools orboardrooms, might thus be able to share a common background ofknowledge with the workplace.My project, then, uses an ethnographically basedmethodology to examine in some detail a collaborativelywritten business proposal. I then use my data to build amodel of a representative case of the writing process in asocial context.18Such a procedure is largely, I believe, withoutprecedent. I have found, however, that a model based uponethnographically gathered data proves useful for a number ofreasons. First, it is based on rich and sensitivelycontextualized data. As I examined and re-examined my videotapes, I began to see levels of engagement with text which Isketched into a diagram. More analysis showed characteristicswhich clustered around each category, characteristics whichallowed me to examine further events and test the reliabilityof the model. As Cerebellum writers found with visualrepresentations, diagrams which help to explain ideas tofellow collaborators are often useful with a wider audience ofreaders. One hopes they will feel welcomed as members of theresearch community. The model also shows data to these othersin a form more condensed than ethnographic narrative or rawtape footage, to allow for ease of discussion by variousaudiences. Rhetorically, the model seems to belong more to acognitively based theory of writing than a socially based one.However, it is my intent to draw together the best ofapproaches to writing in order to extend our ability to assistcomposition in many environments. Business is a field whichvalues the efficient use of models to quickly and clearlyexpress ideas, perhaps since the graphic representation is adominant symbol system in the “technologized context of thetwentieth and twenty-first centuries” (Hampton, 1990, p. 348).19The chronological description of the proposal writingprocess was useful as a way to show some of the socialdynamics between and among writers engaged in text production.In particular, I now argue that their application ofpreviously composed prose may be more widespread, andcertainly more significant to the process than has beenpreviously realized (see, for example Event Q The Recyclersfor the use of already composed diagrams and letters).Although the use of “boilerplates” has long been recognized bybusiness employees and noted in the professional literature(for example Selzer, 1983), I began to see that the use ofpreviously composed ideas was far more complex than the simpleuse of a formula, and thus I called these at least partiallypre-formed ideas “clip-text”. Throughout the collaborationprocess, I also became aware of on-going activities whicheither help or hinder writers working together in a discoursecommunity. I chose to identify collaborators by positionrather than name in order to focus readers’ attention on theirroles (as account executives, or directors, for example)My research will be useful in a number of ways. It willprovide an ethnographically based approach to research byeducators in the workplace. This investigation will usevideo-taping, involving informants in the process, andrepresenting results in a model understandable and readilyusable by employees, teachers, and students. My findings willgive business writing researchers new directions for furtherinvestigation and confirm some previously established findings20on collaborative writing behaviours. Finally, the model oflevels of engagement with text should help to focus furtherresearch which will look at the collaborative writing processin other venues.Upcoming chapters then, deal first with a review of theprofessional literature on collaborative writing in theworkplace. Then, I look closely at my methodology. Data ispresented next in a chronology detailing the development of aproposal at Cerebellum Inc. Finally, I look at conclusionsand implications which are suggested by data analysis andrepresentation in a model. The outcomes outlined above,tentative though they must be, nevertheless, I hope, willinvite you to consider with me the events and ideas thatfollow.21CRAPTER TWO: REVIEW OF PROFESSIONAL LITERATUREBefore I built a wall I’d ask to knowWhat I was walling in or walling out,And to whom I was like to give offence.Something there is that doesn’t love a wall,That wants it down.(Robert Frost, Mending Wall)Although cognitive studies of composition have dominatededucational writing research (Flower and Hayes, 1981;Scardamalia and Bereiter, 1987; Freedman et al., 1987) andcontributed much to our understanding of how the individualproceeds in a composition situation, more socially sensitiveresearch approaches are now enjoying pre-eminence (LeFevre,1989; Flower, 1989; Thralls, 1992) . This is true ininvestigations into academic writing, but is even more obviousin research which examines writing practices in the workplace.The new dominance of context-specific research, which takes asgiven the socially constructed nature of knowledge buildingpractices, has lead to exciting gains in our understanding ofthe patterns of collaboration. Within such a socioculturalparadigm, yet still mindful of the gains of cognitiveresearchers, I place my own investigation. Building a modelof the collaborative writing process, and describing thevarious activities it represents draws together scholarship inwriting from different perspectives.To begin then, it is necessary to look at the study ofwriting in general. In this chapter, I will develop themes22which were introduced briefly in Chapter One. These willinclude cognitive and sociocultural writing research,organizational communication, collaboration and modelbuilding.Overwhelming though the quantity and diversity ofcomposition research may be, scholarship trends (Herrington,1989; Durst, 1990; Fulkerson, 1990; Berkenkotter, 1991) showthat, of several identifiable ways to engage in compositionexploration, two major perspectives are cognitive andsociocultural approaches. Researchers within these twoparadigms approach writing in different ways and,unfortunately, are all too frequently forced into adversarialpostures. Though such behaviour may often be largelycounterproductive to the progress of composition research as awhole, debate continues. It does, at least, challenge newcontributors to defend their place, and brings a multi-vocalquality to composition research which serves to enrich thefield.Given such a struggle, I hesitate to align myself toostrongly with one group or the other, reluctant to give upinsights which are to be gained by a diversity of approaches.Mine is an interdisciplinary study. It seems ironic then,given that I have welcomed the richness found by looking atsuch fields as anthropology, philosophy, and sociology withinmy work, to find that I cannot so freely adopt techniques fromthe cognitivists and the socioculturalists without someexplanation for my decision. Therefore, I will take this23opportunity to develop an argument which seeks to marry someaspects of both the cognitive and the sociocultural approach.Cognitive ResearchAs a educator, I am especially drawn to the exciting workof cognitive psychologists in writing research of the past twodecades. As a composition teacher, I have always hoped thatmy interactions with students will help them to become betterwriters. Defining better is of course a challenge. Still, achange in emphasis from examining products to examiningprocess fostered by cognitive research has helped writingteachers to focus on the specific behaviours which leadwriters to more or less sophisticated products.Emig (1971) is often credited with seminal work in usingcase study methodology, adapted from psychological approachesto research in order to study writing. She examined theactivities engaged in by outstanding grade twelve students (asnominated by department heads) . Though observation conditionswere somewhat artificial (out of a classroom context, forexample) and subjects were asked to perform “think-aloud”protocols, which have since come under some suspicion as aresearch technique (Bizzell, 1982), nevertheless, educatorswelcomed an approach which offered a chance to look at whatwriters were actually doing while they composed. Once studieswere also done with less successful writers, comparisons couldbe made and then, most importantly, we could get on with thebusiness of teaching better approaches to make poorer writers24perform more like good writers with, we hoped, a significantlybetter written product.Cognitive investigations often examine the differencesbetween experts and novices (see, for example, Flower; Flowerand Hayes; Scardamalia and Bereiter; Haas). While interestedin building a theory of writing in general (Flower and Hayes,1981), cognitivists also offered a way to intervene, toimprove the writing of students and make it moresophisticated. Good writers, cognitivists suggested, werebetter at solving problems. These superior writers couldbuild a model of their writing difficulty and find a solutionfor their writing impasse to reach a compositional goal.Better writers were able to take knowledge and not merelyreport it (copying, and not really composing), but rather“transform” knowledge (Scardamalia and Bereiter, 1987) in amodel building fashion. These expert writers were “activebuilders of knowledge”, engaged in “intentional cognition.”They seemed to understand that “the composing processconsists of setting goals, formulating problems, evaluatingdecisions, and planning in the light of prior goals anddecisions” (p. 362) . Such an awareness, coupled with anability to participate in the best procedures for realizingtheir plans, seemed to lead to a superior writing performance.Cognitive research in writing processes has informededucational approaches in the public school system sinceshortly after the appearance of Emig’s 1971 research and itsformalization by Flower and Hayes with their model of the25cognitive writing process in 1981. It has helped teachers tomake informed decisions about what to do in writing classroomswith Grade One students and Grade Twelve students, and withbetter and poorer performers. Cognitive research has alsointroduced the importance of the professional writer as amodel. For example, Pulitzer prize winning author DonaldMurray is often studied in hopes of discovering more about howthe best writers proceed so that educators may assist the lessable.Cognitive approaches to writing in the nonacademic areasare congruent with those found in academic pedagogies. Thosewho write as a part of their career have a greater stake inthe pragmatic outcomes of their ability to write well, so theyperhaps are even better at the skills we would impart to ourschool aged writers:Business and professional writing calls forprofessional-level rhetorical problem solving. Aswith any important act of cognition, when writersknow the real dimensions of the task and theknowledge it calls for, they are more likely tomanage their own writing and thinking with the sameawareness they bring to other aspects of theirprofessional life (Flower, l989b, p. 36)Cognitivists speculated that such an approach, sensitive tothe principles of rhetoric as interpreted from a cognitivepsychological point of view, would lead to improved workplacewriting. For example, Beck (1992) advises technical writersthat26Ebjy considering proposals in a rhetorical context,writers can enhance their proposals by focusing onthe ethos of the situation: highlighting thecompany’s ability and desire to complete therequested task in light of the audience that willevaluate the proposal (p. 125)Each individual writer, it seems, must consider using allrhetorical appeals—ethos, pathos, logos—as ways to solve theproblem suggested by the writing project. As an educator, Ivalue the research done by cognitivists, especially in lookingat procedures followed by the individual writer, interveningin that writer’s approach, and presenting a model of how thewriter proceeds.Recently, the cognitive group has admitted that theirwork would be strengthened by more “sociocognitive” views(Langer, 1985; Freedman, Dyson, Flower, and Chafe, 1987).Cognitive researchers began to discover that students wereinfluenced by home and classroom contexts in their writingpractices, a seeming move to a more socially sensitiveoutlook. However, Flower has forcefully asserted that herattention will remain focused on the thinking and writingabilities of each individual student, rather than becoming tooinvolved in a socially based approach, since she finds that“as an educator, the action I can foster does not go on withina social abstraction or a collective, but in the minds ofindividual writers. The ultimate reason for my research isintervention” (Flower, 1989a, p. 295)27Thus it can be seen that cognitivist researchers proceedin a manner which mirrors their beliefs about writers. Toreiterate, then, according to cognitivists, there is aproblem: poor writing. Rhetorically speaking; this might beprose which fails, for whatever reason, to persuade anaudience to accept an argument. The goal of such research isobvious: good writing, that is, writing which is effective inconvincing a target group. Good writing, however, is but aproduct. In order to achieve a better product, the processmust be examined and evaluated, and then steps taken to allownovice writers to proceed as expert writers would and thusachieve improved production. There is an underlyingpresumption that some rules must exist. For example, expertwriters transform knowledge by doing thus and so, whilenovices simply re-tell knowledge in its original form. Goodwriters are problem solvers: they compose to work outdifficulties by examining alternatives (e.g. this word insteadof that, this argument first rather than last) to effect thebest solution. Such research, however, while offeringdefinite answers to the composition teachers, remains, Isuggest, too entrenched in a remedial mode, and too focused onthe isolated inventor of text to be entirely useful in mypresent work.28Sociocultural ResearchA more sociocognitive view accepts that the rules ofcomposition, as used by individual writers, are affected bythe groups in which these writers work. Interest in thefunctioning of the individual within a group of writers, nowoften called a “discourse community,” has lead gradually to ashift in focus by some composition researchers. Theseresearchers suggest that our real interest ought to be focusedon the dynamics of the social group as it struggles to buildlanguage and to build knowledge concomitantly. Many sociallyfocused investigators seek to separate themselves from acontextually influenced view of the problem solvingindividual. They refuse to embrace both views: “Thedifference between saying that language has a social contextand that language is a social construct defines a keydifference between cognitive and social constructionist workin composition” (Bruffee, 1986, p. 784) . Such an assertion,it seems, leads us to a consideration of truth, as built forexample by a writer or group of writers, as a relative, non-positivistic concept. Those collaborators, building meaningand text together, seem indeed to be constructing ideastogether rather than merely doing writing in isolation. Toinvestigate the difference between a more cognitive and moresociocultural view of writing, however, it will first behelpful to consider the debate surrounding two key terms inthis research controversy: audience and community.29Audiences and CommunitiesAristotle’s view of listeners/readers clearly places themin a role of commensurate importance with speakers andwriters. Recently, however, attention to audience, certainlyin the case of social constructionists, is being challenged byanother concept: that of “discourse community.” Some tensionexists between applications of these terms (Bizzell, 1982)However, this controversy should not blind us to thesimilarities of classical views of audience and the morecontemporary stand of social constructionists that discoursecommunity is of greater importance. While concepts ofaudience and discourse community are not identical, they offermany of the same insights which describe the idealrelationship of the speaker and the one(s) spoken to. Somefurther explication of the perceived contrast between audienceand community will show the need for attention to this topicin developing further understanding of the collaborativehabits of writers.Aristotle views rhetoric as not only a psychologicalpursuit, but a.sociological one as well. Cooper (1932) saysof Aristotle’s Rhetoric that it is ‘ta searching study of theaudience ... the speaker or writer must know the nature of thesoul he wishes to persuade ... it thus becomes a populartreatise on the interests of men in groups and as individuals”(p. xx) . Aristotle believes, for example, that the rhetor’scharacter “is the most potent of all the means to persuasion”(p. 9). The speaker adapts to the speech. Aristotle cautions30the rhetor who does not study the audience carefully. Such aspeaker’s argument will be “unconvincing because theconclusions are drawn from premises that are not admitted orcommonly believed” (p. 12). The rhetor discovers groupbeliefs, and thus is able to construct an appropriateargument. Though Aristotle does, for example, set the rhetorabove the audience in intellectual ability, as the audiencemay be “uncultivated” (p. 154) or the “sort of hearers whocannot grasp many points in a single view, or follow a longchain of reasoning” (p. 11), the success of the speaker isstill based on a sensitivity to the beliefs of the communitywhich the rhetor joins when making a speech.Modern attention to Aristotle’s ideas (Perelman, 1982)finds that his premises are still valid: “to make hisdiscourse effective, a speaker must adapt to his audiencethe speaker can choose as his points of departure only thetheses accepted by those he addresses” (p. 21) . Perelmanfinds that arguments become more effective as they arerendered more commensurate with an audience’s convictions andtraditions (p. 140); however, these arguments are notmanipulative. Instead, the discourse offered “tries to gain ameeting of minds instead of imposing its will throughconstraint or conditioning ... a meeting which social andpolitical institutions can facilitate or prevent” (p. 11)Similarly, Mao (1989) observes that in such meetings, thereare no triumphant speakers and conquered listeners, but rather“everybody wins” and “persuasive discourse ... [becomes] part31of a continuous cooperative effort to identify what can bebest shared by writer and audience” (p. 139).Thus we see that, while attention to classicalconceptions of audience are valuable for an emphasis on thecognitive functioning of the individual composer, so too arethey evocative within a more socially focused framework.Late twentieth century social constructionists, who understand“knowledge and the authority of knowledge as community-generated, community-maintaining symbolic artifacts” (Bruf fee,1986, p. 777), can still believe in an Aristotelean importanceof the behaviour of each individual within a collectivebecause: “[E]thos arises from the relationship between theindividual and the community (LeFevre, 1987, p. 45)”.Some reluctance to accept such a view still remains inthe social constructionist group. These scholars are notcertain that the concept of audience is helpful in our effortsto understand the workings of the community. If “[t]hecommunity of knowledgeable peers constituted by that symbolsystem [language] constructs knowledge by justifying itsocially, that is, by arriving at a sort of consensus”(Bruf fee, 1986, p. 779) then perhaps, they argue, it is timeto dispense with audience and its unappealing baggage ofseparate realities between speaker and listener:A social perspective on writing creates the need fora new metaphor, one which can suggest the richsocial dynamics that surround and support text, onewhich can account for the reciprocal relationshipsamong writers and readers described by recent32research of the writers and readers within them(Pare, 1991, P. 49)Efforts have been made to recast audience as a more fluidand more flexible concept than the oft-perceived adversarialone (Ede and Lunsford, 1984; Kroll, 1984; Roth, 1987).However, the audience concept is still seen by many as atleast unhelpful (Elbow, 1987) and at worst restrictive andinaccurate because, some social constructionists argue: tithetext is not a message delivered to an ‘audience’; rather, itis a moment in an ongoing discussion, an utterance shaped bythe relationships, concerns, and procedures of the community”(Pare, 1991, p. 51)For these reasons, social constructionists find the ideaof the discourse community, that is “a group sharing languageusing practices” (Bizzell, 1987, p. 1), as a more usefulconcept to describe the relationships between readers andwriters, speakers and listeners. Social constructionists findthat their activities are based on a view that “[w)e uselanguage primarily to join communities we do not yet belong toand to cement our membership in communities we already belongto” (Bruf fee, 1986, p. 784). Rather than emphasizing thediscovery of universal rules which the novice might use tobecome an expert, a sociocultural stance sees that the writeris persuaded to shape writing behaviour to match the style ofthe target community.Discourse communities, although somewhat difficult todefine (see Harris, 1989), seem to be bound together by a33discursive project. Such interpretive activities, which mustbe undertaken in accomplishing the task, delimit the group anddraw it together. Therefore, “some work in the world itsmembers could not accomplish on their own” (Bizzell, 1987, p.1) may be successfully done collectively. These discoursecommunities, in turn, influence the members who belong tothem. Indeed, the group finds itself controlled by variousrules of conduct which the group has helped to determine,whether individuals are aware of such codification or not.Thereafter, “the norms define the writers discoursecommunity, a context that conditions, governs, and constrains,not just the message, but the writer producing it” (Freed andBroadhead, 1987, p. 162).The pedagogical implications of examining classrooms asplaces to learn about how to join other discourse communitieswere obvious to social constructionists. Rather than lookingat universal rules for better writing as cognitivists might,“composition studies should focus upon practice withininterpretive communities——exactly how conventions work in theworld and how they are transmitted” (Bizzell, 1982, p. 239).In such a discourse community, students would be viewed not asdeficient writers as the cognitivists seemed to imply, butrather as writers unknowledgeable about the habits of aparticular discourse community which they sought to join (forexample, the community of literary analysts).34Thus, the student gains not a set of universal rules butrather learns that even knowledge and skills attained in theclassroom are subject to change. Nevertheless, educatorsmight still teach discourse analysis, to discover the ruleswhich seem to govern the organization and style of a text fora particular group, in order to allow students easier accessto the many discourse groups they might wish to join.Students might also be able to achieve, as Harris (1989)suggests, a kind of polyphony “not to initiate our studentsinto the values and practices of some new community, but tooffer them the chance to reflect critically on thosediscourses. . .to which they already belong” (p. 19).In similar fashion, business or professional groups ofwriters, which the students may want to emulate, are involvedin creating their own culture through various sorts oflanguaging. Kuhn (1970) notes that “knowledge, like language,is intrinsically the common property of a group or elsenothing at all. To understand it we shall need to know thespecial characteristics of the groups that create and use it”(p. 210) . One sort of common property of a discourse group,especially in business, seems to be a store of previouslywritten texts from which writers can draw ideas, formats, andeven whole sections. Academic writers too find that they areexpected to build upon each other’s contributions to thecommunity, and they too use words and sentences taken inchunks from the texts of others (or indeed from their ownprevious papers) . Each discourse community, it seems, has a35kind of data base from which it draws “clip-text” in much thesame way as publishers and others use “clip-art.” Thedifferences seem to lie in the extent to which the clippingsare referenced and transformed. Certainly, boilerplating, orwhat Bereiter and Scardamalia (1987) might call “knowledgetelling” can be a problematic activity and is identified assuch by some business writing scholars (see, for example,Wallace, 1994); however it is widespread enough in theworkplace to warrant further consideration.Early conceptions of discourse communities sought tobreak away from a focus on the individual writer addressing anaudience, preferring to consider language’s social functions(for example, Vygotsky, 1978; Halliday, 1969) and thereforethe primacy of a social focus in research: “Conceivedtraditionally as an individualizing and adversarialrelationship, writing viewed as a form of instrumental speechbecomes a referential and interdependent one ... writingbecomes essentially and inextricably social or collaborativein nature” (Bruf fee, 1983, p. 165) . While such a view seemsat least to admit the possibility of divergent points of view,even if such points are of lesser importance, other sociallymotivated perspectives are less tolerant. Even ifcognitivists, for example, were willing to admit a contextualinfluence, some social constructionists seem definitely lesswilling to find middle ground:[Hiuman language (including writing) can beunderstood only from the perspective of a society36rather than a single individual ... the focus of asocial view of writing, therefore, is not on how thesocial situation influences the individual, but onhow the individual is a constituent of a culture(Faigley, 1986, P. 535)One of the main goals of the social constructionistcommunity thus came to be the study of how the discoursecommunity participates in epistemological ventures. While onemay argue that they failed to grant that the cognitivists,though engaged in the same project as themselves (discoveringhow writing proceeds), might be working under different normsand therefore quite likely to come to different conclusions,the social constructionists nevertheless sought to bothdifferentiate themselves and prove that they espoused thesuperior approach. They maintained that theirs was a realcommunity, as they defined it:Social construction understands reality, knowledge,thought, facts, texts, selves, and so on ascommunity-generated and community-maintainedlinguistic entities--or, more broadly speaking,symbolic entities--that define or “constitute” thecommunities that generate them (Bruf fee, 1986, p.776)Communities were not primarily concerned with or engaged inthe persuasion of some “other” audience, but rather in workingtogether and persuading themselves. Unlike the cognitivists,social constructionists assume that “there is no such thing asa universal foundation, ground, framework, or structure ofknowledge. There is only an agreement, a consensus arrived atfor the time being by communities of knowledgeable peers”37(Bruffee, 1986, p. 774) . Knowledge is built by communitiescomposed of experts as a kind of “best guess,” but must beperceived as ephemeral.Despite the appeal of the notion of discourse communityfor me in my research into collaborative manners of writing,there can be no doubt that the concept of “audience,” withlong roots in rhetorical history, established and nurtured foralmost two thousand years, remains a powerful idea. Theconcept of audience is undergoing renewal and recasting.However, I suggest, it is far from being, as Pare (1991)maintains, in danger or need of being “ushered out.” Thenineties have seen several attempts to rethink audience. Itis interesting to note that not only do these new attempts tore-position audience help us to think more comprehensivelyabout the readers of our work, but also about the nature ofdiscourse communities which we may share, or come to share, asa result of our rhetorical act.One such recasting is that which imagines audiencemembers as a subgroup of a communicatively enabling community:A discourse community, however it is definedessentially denotes the setting or culture thatenables communication within it ... [it is] thesystem of rules, conventions, constraints, andbeliefs that readers and writers share and draw uponduring the process of communication ... [P]articularaudiences ... nearly always exist within a discoursecommunity (Selzer, 1992, p. 172)Thus, understanding the community’s ways of knowingenables us to address some group of its members more38effectively. If this group is known, for example, to acceptwriting as a communicative act, as an act which looks forinteraction between speaker and bespoken, then speakers andwriters might seek to address their audience according totheir assumptions about their views of valuable discourse:“Good audience analysis thus involves directly the presumptionof writing as a social act in a discourse community”(Fulkerson, 1990, p. 417)A second recasting seeks to show how the communitydefines the activity of audience analysis. As Kirsch (1991)discovered, within one group, individual members differ intheir approach to addressing other members of that same group.She maintains that her researchcontributes to the social-constructionist inquiry byfocusing on writers’ sensitivity to social contextsfor written communication and by tracing individualdifferences--as well as similarities--in writers’sense of audience ... how writers representaudiences which occupy a different sociopoliticalstatus within a given community (p. 34).Such research also rejects any notion of homogeneity within adiscourse community, and reminds us that the individual writerstill makes individual decisions, despite the force of thegroup’s influence. Kirsch does not however, look at theinteractions of collaborators who seek to represent audienceswho may be outside their community.A definition of discourse community, expanded toencompass the often varying activities and opinions, must deal39not only with the idea of audience but also with sometroubling assumptions about the demeanour of the group. Itis, for example, commonly assumed that communities are placesof calm, where people join together for a common good and,therefore perhaps, are quick to arrive at a consensus. ThoughBurke (1957) assures us that his famous parlour is a forum ofdebate, one cannot somehow, especially given his premise ofthe identification desirable between conversants, conceive ofa major conflict in his parlour. However, other scholarsremind us that although every community has a “‘core’ ofcommonality” (Lyon, 1992, p. 286), just as important is thecommunity’s inherent “turbulence of action” (p. 286). Anyrhetorical theory which fails to note the importance ofdissent, the discourse which does not meet group standardswhich may lead to breakthroughs in the group’s understandingof an issue under debate (Rorty’s abnormal discourse), seemsto me to be in danger of denying the inherent dynamism oflanguage’s social functions.In practice, the discourse community of the schoolroommust deal with the results of the ubiquitous (and often badlyimplemented) group project which too often results not in thestruggle of ideas, but rather to a sullen lowering ofdiscussion to a superficial, yet consensual, commondenominator: “Any teacher who uses group discussions orprojects has seen that they can, on occasion, be fierceenforcers of conformity” (Myers, 1986, p. 159). Ewald andMacCallum (1990) offer a plan for a classroom community which40welcomes dissenting opinions. Their pedagogy includesencouraging “deferral of agreement [which] becomes a creativeand, therefore, desirable group strategy” (p. 23). This“rhetoric of dissensus” (Trimbur, 1989) should certainlyrevitalize our conceptions of the community to look at thegroup, and at the skills of the individuals who bothconstitute and are constituted by the community.John Dewey (1938) maintained that unproductive argumentsmay result from raising a wall between the individual andsociety: a false dichotomy. Educators, he believed, would dowell to remember that “individuals are parts of a community,not outside of it” (p. 54). Similar ideas are expressed bymore recent scholars. The discourse community must include arecognition of the group and of the individual members: “Wewill more fully comprehend the process of creating new ideaswhen we think of it as an act that is social even as it isindividual, with the other always implicated in the inventionsof the I” (LeFevre, 1987, p. 140) . In my case study andpresentation of the model, I have endeavured to do just that.Organizational CommunicationWriting in business develops in an often treacherous seaof group dynamics: the unspoken assumptions, traditional rolesand cultural norms which swirl like eddies in a stream, seenand unseen, helpful and unhelpful to the completion of thetask. And indeed, the product is paramount and primary in thework setting. Employees writing together are deployed41according to their ability to contribute to the finaldocument. In the educational setting, by contrast, the mainconcern is that students learn together through contributingequally to a task and being co-authors (Dauite, 1986) in acompositional process. Any practising teacher, however, willtestify that students often resist such egalitarian models.Nevertheless, the classroom attempts at least to providesimilar opportunities whereas the business world proceeds in amore pragmatic (and some may say even Machaviellian) fashion.Business, nevertheless, is still concerned withmaximizing the effectiveness of the task group. To this end,group members must deal with centrifugal and centripetalforces of collaboration in a productive fashion. Problem-solving groups, such as those charged with writing a proposal“have in common their ephemeral nature and their dedication toone ... task, and the task will usually have an ending point,that is, a deadline” (Malone, 1991, p. 110). The group musttherefore quickly identify the roles of group members: “acontributor must have a clear view of what all members of theteam see as the project’s elements and constraints” (Newman,1988, p. 37), and accept them, at least to some extent, inorder to proceed.Indeed, such building of and adherence to group normsoften comes in the first meeting, and patterns, bothbeneficial and detrimental, tend to continue throughout thegroup’s brief life as a discourse community. However, aschallenges to the speedy and effective completion of the task42are faced and require some recursion in the process,individuals within the group must be prepared to function awayfrom group norms to adapt to changes (Malone, 1991),recognizing, for example, that the problem as defined cannotbe solved and must therefore be re-defined. Individuals aswell as groups, however, are frequently inclined to persistwith an established plan, even when there is strong evidencethat they are pursuing a losing course of action (Staw andRoss, 1989)A concern with collaboration among group members is anatural outcome of the small group dynamic process. Subgroups must also work toward a satisfactory conclusion of themain project. Ideally, the best maintenance of creativetension between the centrifugal and centripetal forces wouldlead to situations where “all collaborators might be convincedthat they could maintain self-identity and still bond withgroup members” (Lay, 1992, p. 91). Although it has beensuggested that the most effective communicators are shapedeven more by past experiences than by gender (Tebeaux, 1990),business writing in particular seems to be best served by amix of styles and approaches.Taking time to build social relationships beforebeginning work on a project requiring collaboration (Varner,1988) helps to build teams capable of both entering andwithstanding a communal frame of mind. Experience in previouscollaborations with the same people, for example, seems tohelp group members engage more successfully in present43collaborations (Allen et al., 1987). Not many businesses,however, have the time to build such relationships. Groupsmay be formed from a mix of employees who were hired to do aparticular job and will be dismissed or moved to anotherbranch office, when they are finished, to begin anotherproject with yet another group. Modern businesses are, thus,especially challenged by the maintenance of effectiveorganizational communication.Collaborative WritingThough research into the nature of collaborationcontinues apace, definitions of collaborative writing for bothacademic and nonacademic writing remain elusive. Those whichdo exist seem most often to be shaped context by context. Forexample, it is hard to deny that in the broadest theoreticalsense, all text arises from others, other texts by otherwriters and other ideas by other speakers: “[Wiriting iscollaboration. It cannot be otherwise” (Reither and Vipond,1989, p. 866) . Such a concept, that all text production iscollaborative, uncomfortable though it may be for those whocontinue to demand a place for individual agency (and evengenius), is difficult to avoid in business composition. Inthe workplace, many documents are produced by teams, and evenindividuals consult constantly with previous documents andcompany “boilerplates”, other employees, and the clientaudience in the building of text. It seems, in corporateCanada at least, that “collaboration 1is1 a partnership44present in all discourse-production situations” (Thralls,1992, p. 65) . Creation of new text involves others ascollaborators: frequently “people interact to invent and tocreate a resonating environment for inventors” (LeFevre, 1987,p. 50). These people may presumably be absent in time andplace.True as such pronouncements might be in a general sense,however, such collaborative partnerships ignore time andchoice. In this document I draw upon Aristotle to help mebuild my text. He had no choice in what I create. However, Ido my composition in his debt. He is also, of course, absentin time and place. Ede and Lunsford, while my contemporaries,have also no choice in my use of their ideas. However, myconsciousness of the opinions of the composition community inwhich I claim membership constrains my words: I build mydiscourse carefully on my understanding of reasonableconclusions based on all of my absent “collaborators”. Thus,I am aware of the social “collective” (LeFevre, 1987) as: “asupra-individual entity ... encouraged or constrained byinstitutions, societal prohibitions, and culturalexpectations” (p. 50)In some other, perhaps more practical, certainly more“here and now” sense, Aristotle, Lunsford and Ede are not mycollaborators at all. They have no stake in my dissertation.Even more certainly, in the business writing community,when we talk of collaborative writing, we assumethat more than one person is responsible for a45document. ... [Slurveys tell us that the writingprocess in industry is collaborative if we take theword in its broadest sense to mean “workingtogether” (Debs, 1991, p. 478)Within this more restricted sense, that is, “writing in whichmore than one person contributes to the effort” (Couture andRymer, 1989, p. 73), collaboration remains somewhat difficultto define but easier to recognize. Two or more peoplelabouring together to some common purpose is indeed “the kindof beast we know when we see it” (Debs, 1991, p. 479)More specific definitions, while valuable within acertain specific context, become open to attack if takenoutside their immediate situation. Care must be taken to bothbroadly situate collaborative activity and recognize itspresent incarnation. One example may be seen in a descriptionof collaboration as “a variety of interactive writingexperiences” which in one specific case becomes “collaboratorsproducing a shared document, engaging in substantiveinteraction about that document, and sharing decision-makingpower and responsibility for it” (Allen et al., 1987). Suchdefinitions may, however, lead to further debate about nonspecific terms such as “substantive” (such as, for example:are less substantive interactions less collaborative, or evencollaborative at all?) I have, therefore, adopted a broaderdefinition which will have less power to disallow one “workingtogether” situation over another as suitable for research.Collaboration researchers must be prepared to deal withsituations of ambiguity. Some investigators lament that there46remains “lack of agreement about the term, [which results in]diverse commentary on a destabilized concept and practicesrather than an agreed upon, cumulative body of knowledge basedupon a theoretically unified position” (Forman, 1991, p. 235).Such theory building, however, cannot proceed unlessresearchers strive to describe, whether by definition as somedo, or by model, as I have, the manners of collaboration forany situation, while keeping in mind the broader meanings ofthe term.Specific Research FindingsKnowledge building of a theory of collaborative writingin business rests largely on data accumulated from surveys,interviews and questionnaires. Early work centred onquestions of whether writers did indeed collaborate, whichseems now to be irrefutable (Faigley and Miller, 1982;Anderson, 1985; Ede and Lunsford, 1986; 1990) . Inquestionnaires which surveyed hundreds of employees, writersstudied by these researchers reported that large percentagesof them “sometimes collaborate with at least one other personin writing” (Faigley and Miller, 1982) . However, academicsconducting such research began to realize that the definitionof collaborative writing might vary between the academic andnonacademic communities. Such confusion, arising from anacademic notion of collaboration among equal partners, mightbe at variance with a business view: “results indicating thatcollaboration is typical on the job may be misleading”47(Couture and Rymer, 1989, P. 75). Some jobs in the workplaceseem to call more regularly for collaboration. For example,those requiring quality as a factor more important than, or atleast equal to, speed of production. A large percentage ofthese writing tasks were completed by writers working together(76%). More routine jobs, in contrast, were often completedalone (Couture and Rymer, 1989).Collaborative writing projects were also viewed bywriters with decidedly mixed feelings. Research techniquessuch as interviews revealed that workers might prefer to workalone but saw the practical results of group approaches.Although individuals saw sacrifices of time and ego, and evenethical stance (Bryan, 1992), they also appreciated that,effectively managed, groups could offer a better product.Collaborators also understood that conflicts could lead tomore creative solutions; large projects could be dividedamongst several writers; specialists could contribute highquality input; and teams could synthesize large quantities ofinformation more effectively than individuals could (Allen etal., 1987).More recently, research into collaborative processes hasmoved beyond the tabulation of the numbers of writers involvedin writing to a consideration of the patterns revealed inmanners of collaboration. Ede and Lunsford (1990) havepublished a large study which reveals widespread collaborationof different types. They identify and discuss two modes ofcollaborating, the more widespread “hierarchical” (one member48of the collaborative group having more power, others less) andthe less frequent “dialogic” mode (all members having equalpower) . However, they resist setting up modes asbinaryopposites (or as negative and positive modes) given the natureof their survey, questionnaire, and interview data: “Perhapsonly full-fledged ethnographic studies could provide the depthof detail and critical perspective necessary... Discoursesituations are ... inherently mixed and paradoxical; they defyeasy analysis and categorization” (p. 134) . And indeed,ethnographic studies of business writing do exist. Cross(1990), using the work of Bakhtin, analyzes her data andidentifies sixteen factors which affect the success ofcollaboration based on notions of small group cohesion. Thesefactors are centripetal ones such as time, centrifugal onessuch as differing perceptions of audience. Still otherfactors, such as a group hierarchy, are convertible; that is,they could contribute to the building of ties between groupmembers or tend to drive them apart.Both Doheny-Farina (1986) and Brown and Herndl (1986)discovered through their ethnographies the importance of theenvironment where the collaborative writing process takesplace. Doheny-Farina suggests a model which describes areciprocal relationship between writing and organizationalcontext. He postulates that context influences the waywriters formulate their situation, and the way they behavewithin that situation. In turn, the activities of thesewriters influence the organizational context. Brown and49Herndl add that the context which surrounds the social writingprocess creates a lack of privacy or solitude; rather,“professional writing is often display behaviour in a verypublic place” (p. 24). Writers in corporations “describ[e) a‘layered’ speech event in which the writer addresses anaudience ... and the entire transaction is observed as atransaction by a known or imagined overlooking ‘other’” (pp.24-25). Furthermore, the writing behaviours engaged in bybusiness writers indicate that they want to align themselvesto a group, and hope that their writing style will mark themas group members and entitle them to group support.The influence of the culture surrounding the writer wasalso investigated by Mclsaac and Aschauer (1990). Theyconducted interviews, circulated questionnaires and diddiscourse analyses of engineers’ writing. They sought tocharacterize the proposal writing process and discovered ahighly collaborative, highly structured approach to thewriting task. Further, they discovered that collaborativeassistance helped improve the quality of the proposal andinfluenced writers’ attitude towards writing. They concludedthat “[ilanguage practices are rooted in an organization’sculture” (p. 551) and recommended the importance of personalinteraction to improve writing. Their research is echoed byadvice from various business communications experts teachingseminars in the workplace (Asner, 1991; Harcourt, 1990).50Questions RemainingMy own investigation seeks to address some gaps whichpresently exist in knowledge about the collaborative writingprocess in the workplace. In addition to calls for moredetailed observation of business writers CEde and Lunsford,1990), gaps exist also in our understanding ofthe extent to which workplace writing is immersed in‘talk’ and the nature of these exchanges. ... [M)orecase studies could help define the interactions thatoccur while planning and drafting are in process,demonstrating, for example, how closely some pairsof writers collaborate and whether their sense ofauthorship becomes merged (Couture and Rymer, 1989).Such studies should, of course, look also at more problematicinteractions, such as the problems faced by groups whichinclude reluctant collaborators (Stohi and Schell, 1991), andthe difficulties of developing a collaborative representationof audience.In addition, work must be done in focusing on the variousactivities which surround the creation of discourse, byexamining “the relationship of a text to other texts and toconversations that precede and follow it ... [and) the writer’srelationships to other people ... who affect invention”(LeFevre, 1987, p. 125) . Such investigations need to findways to represent the complex contexts which influence thesocial writing process. One of the questions arising in theworkplace is the nature and prevalence of previously shapedideas. Selzer (1983) notes that his subject, Nelson, “often51borrows sentences, paragraphs, sections--even graphs--frompast documents and incorporates them into new proposals,reports, and correspondence” (p. 181). In similar fashion,Winsor (1990) observes that her writer, Phillips, spent muchof his time “writing from previous texts” (p. 61). However,Winsor finds also that Phillips interprets existing materialas he writes new text. Both Nelson and Phillips are observedwriting in isolation; that is, not as a part of acollaborative effort. They are also both acknowledged bytheir respective researchers as talented writers who evensought out writing tasks. We need to look at the “clipping”of previously composed ideas and texts by collaborators whowere often more reluctant writers.Next, we need to continue to address the difficulty ofdefining collaboration, both in our own community and in othercultures such as those in the workplace. LeFevre comments:“Writers in all discourse communities should attempt toclarify and possibly expand the prevailing sense of whatconstitutes collaboration and how it should be acknowledged”(p.123) . As well, work needs to be done on ways to share ourfindings with an academic research community and with thebusiness community. Such approaches should include moreinvolvement of informants as participants in research.We need to develop representations which allow us toreveal our growing understanding of the rules which seem togovern collaborative activities, whether these representationsare models (Smudde, 1991; Nystrand, 1989) or modes (Ede and52Lunsford, 1990) . All such attempts are bound to becontroversial, revealing as they do an attempt to makeconcrete what is obviously abstract, and simple what isremarkably complex. The advantages of a representation aremany as well, however. A model is a visual representationwhich offers a distilled and, it is to be hoped, potentversion of a researcher’s findings. “Maybe the social writingprocess looks like this,” the researcher says, invitingfurther conversation with the model as focal point. Such arepresentation must therefore be sufficiently but notoverwhelmingly complex, must be convincingly supported bydata, and finally must be useful in engendering furtherresearch and scholarly debate.Despite the risks of modelling which is too specific ortoo general, researchers are already inclined to developconceptual frameworks because “[tjhey assist us in makingsense of the flood of sensory data we receive every wakingmoment from the world around us, and they help us tosystematize that world by revealing its underlying patternsand regularities” (Pemberton, 1993) . Attempts to developmodels such as that by Flower and Hayes (1981) to describe theindividual’s cognitive writing process were followed by aflurry of research activity suggested by the model. Eventhough the Flower and Hayes model has been sharply criticized(for example, Bizzell, 1982) and more recently found toolimited even by one of its creators (Flower, 1989), it stillsucceeded in giving definition to a complex process. Models53do exist as ways to present social writing processes, butclearly more research is needed to propose a representationwhich will be helpful in extending our work on the activitiesand interactions of collaborative writers. It is to such workin examining and representing collaboration, then, that myresearch is directed.54CHAPTER 3: METHODOLOGYAlice could not help her lips curling up into asmile as she began: “Do you know, I always thoughtUnicorns were fabulous monsters, too! I never sawone alive before!”“Well, now that we have seen each other,” saidthe Unicorn, “if you’ll believe in me, I’ll believein you. Is that a bargain?” (Lewis Carroll, Throughthe Looking Glass, and what Alice found there)In studying the writing habits of another culture, weoften discover that our problems are much like those Alicefaces. When she steps behind the looking glass, Alice findsthings not just reversed, but unusual in many ways. Here, forexample, Alice meets a unicorn whom she finds to be assceptical of her existence as she is of his. Eventually,however, Alice and the Unicorn agree to trust to their senses,ignore old ideas and believe in each other. The Unicorn isbut one of many fantastic characters with whom Alice learns tointeract as she journeys down the rabbit hole, and behind themirror: the Red Queen, the Mad Hatter, Humpty Dumpty, theCheshire Cat, the White Rabbit, Tweedledum and Tweedledee.Their only similarity seems to be their altogether “curious”outlook on life, or rather on life as Alice understands it.They are involved in a number of strange activities, such asunbirthday parties and fencing with umbrellas, and severalmore familiar ones taken to unfamiliar extremes, such asinterpreting “nonsense words” and battling for crowns. Whatis Alice to make of Wonderland and the world beyond thelooking glass?55Alice has several approaches which seem sensible toanyone who seeks to make the strange familiar—as ethnographicresearchers do. She becomes a watcher, a listener, and aparticipant. She talks to the residents and tries to imposesome order on her findings, checking her understanding byengaging in conversation with those she meets. Aliceperseveres, even though her attempts at unraveling thebehaviour of Wonderland’s denizens sometimes make her cry,“Nonsense!” in frustration. Mostly, however, Alice sees herproblems in understanding as a wonderful adventure, especiallyin retrospect as she shares her experiences with her sisterand her cat.If we want to investigate writing activities in anunfamiliar context, we will encounter what may seem to becurious people engaged in curious games, with somesimilarities to the games of our own academic world and alsosome remarkable differences. How can we gain access to andstudy such a world—and how represent it to those members ofour own culture “back home,” many of whoth will be rather morelike the sceptical Unicorn than the accepting cat?In this chapter, I will outline my research approach. Myadventure began in accessing a research location. Next, Iwill describe my activities in gathering data at the researchsite in interaction with those who worked within the culture Idiscovered. Third, I will clarify the techniques used to sortthe data collected, and then, fourth, I will explain themethod used to analyze the sorted data. Finally, I will56discuss my representation of the data for further debate anduse by interested parties within the academy and theinvestigated community.Although I began this investigation as an interpretiveethnographer, attempting primarily to “generate insights,explain events, and to seek understanding” (Anderson, 1989, p.253) of another culture, such understanding did not entirelymeet my evolving research purposes. These purposes wererelationship-building, pragmatic information sharing andeffective writing intervention. Each researcher is influencedby certain research traditions, such as, in my case, those ofethnography, which seems perfectly suited to studying complexsocial phenomena. Though we must be “appropriately but notexcessively suspicious” (Tchudi and Mitchell, 1989, p. 396) ofour conclusions concerning what meaning our subjects may placeon events, we should also question our own culturalassumptions as researchers.Like Alice and the Unicorn, as researcher and researched,my informants and I could have remained locked in twosolitudes. quest to build a relationship between us, Ifound it necessary to push the boundaries of generallyaccepted practices in the gathering, and representing ofethnographic data. I had already moved beyond the confines ofinterpretive ethnography, seeking to understand and influencemy own community as well as the target culture, in order “todescribe and understand the worlds in which researchers aswell as the writers they study reside” (Brodkey, 1987, p.57413). To gather information, I did at first use field notes,but largely abandoned them in favour of video-recording forreasons I will shortly explain. Finally, I realized that,although I had used video to record and gather data in anethnographic fashion, I wanted to represent my findings in amanner other than by ethnographic narrative. Instead, dueprimarily to my business audience, I wanted to attempt topersuade and instruct my readers in both cultures byconstructing a model which would represent the collaborativeactivities of business writers.Gaining Access to DataOf the many groups that researchers might study, the highpowered world of corporate Canada, in my case “Cerebellum” acomputer systems integration company, is a world whichcertainly leads to a questioning of the universityresearcher’s right to assign meaning to events or assume otherhegemonic control. The corporation is, first of all, aculture sometimes at odds with the academy. The oppositionsare easy, perhaps too easy, to enumerate: business is rightwing; the university leans left. Business is pragmatic; theuniversity, idealistic. Business makes money; the universityspends it.Gwendolyn Etter-Lewis, however, reminds us that “[wiemust cease to view the world around us in terms of dualityand/or opposing pairs” (1991, p. 56) but should rather lookfor the possibilities for relationships. The business of58integrating computer systems exudes power, and should lead usaway from techniques which suggest the university’s place issomehow above the life of the corporation. Here we aredealing with a different world, a different culture, but thisworld is quite secure in its wealth and influence. Studyingbusiness writers requires research approaches adapted to theirown special needs--for pragmatism, efficiency, and immediacy.Business people, relate to others on the basis of mutualadvantage, and therefore, research methods used in theirenvironment must similarly seek to provide that advantage.Accessing The Research SiteIn order to pursue a general interest in the workinghabits of business writers, in August of 1991 I became aresearch associate on the Learning Connections Project.Supported and funded by the Social Sciences and HumanitiesResearch Council of Canada (SSHRC), the Learning ConnectionsProject was conceived and pursued by Dr. John Willinsky,University of British Columbia, and Dr. Lori Neilsen, MountSt. Vincent University, Nova Scotia. Their investigationsought to increase an understanding of writing in theworkplace and the schoolroom. They proposed “a university andcorporate partnership that seeks to improve literacy inworkplaces and school by using technologies that help peoplecommunicate” (Neilsen and Willinsky, 1990). Businessemployees at Cerebellum, Inc. and a group of high schoolstudents were introduced to each other and exchanged ideas viaInternet. The students also communicated electronically with59other students across Canada (e.g. between Vancouver, BC andBridgewater, NS) . The interests of the principal investigatorscentred on the influences of technology on communication, andthe influence of correspondents on each other. It was hopedthat such an exchange might lead to gains by all participants,those in school settings and those in business ones.Nearly thirty (out of a total of sixty-five) of theemployees at Cerebellum, Inc. took part in the LearningConnections Project, both by corresponding electronically withstudents (later meeting them face to face) and by submittingcopies of their other, more specifically job related, writtenwork. Over the two years of the project, though studentsubjects remained relatively constant, employee participantscame and went: some hired and electing to join the project;some fired or resigned from the company, some just steppingof f the project. Fortunately, a strong core of participantsremained with Learning Connections through its entire life.Others worked with us very willingly for short periods oftime, for example to fill a spot vacated by an employeerelocated to another branch office. Learning Connectionscontinued despite these often unstable conditions. Theproject team from the university learned to work with a moderncorporation which sees such change as a mostly desirable factof life.Even the highest levels of this branch office ofCerebellum Inc. saw considerable change during the project.Cerebellum has no official hierarchy but a definite sense of60seniority and authority which belies a flat corporatestructure. Shortly after we began, the general manager whofirst allowed us permission to do research was moved from hisposition by corporate head office decision. The new generalmanager, also a company vice president, allowed researchaccess to the employees, but was not willing to be a moreactive member of the project.Accessing the Lives of InformantsAccess, however, is more than just gaining permission toenter a research site. It is also gaining the trust andunderstanding of the participants who may elect to shareopenly or more begrudgingly. Due to the long term nature ofthe project, in my position of business liaison within theproject team, I was able to visit Cerebellum twice weekly andgradually establish long and continuing relationships withmany of the project’s business participants. In addition tomy duties as trouble-shooter during the e-mail correspondence(for example, dealing with problematic messages—incomplete,inchoate, or inappropriate), I drank coffee with employees,went with them for lunch and out shopping, celebrated newbabies (and new photocopiers), welcomed new employees, andcommiserated over job frustrations. I bought flowers duringSecretaries’ Week and was treated more and more as “one of thecompany family,” if a rather itinerant member. My nameappeared on the company employee list and I was invited toplay softball and go bowling. I was assigned a desk,sometimes in the executive wing (more often in the programming61cubicles), and always had access to a private telephone.Curiously, as Alice might have said, in a branch office ofCanada’s largest computer systems integrator, I almost neverhad an assigned desk with a computer (though many employeesinvited me into their work spaces when I needed a few minuteson a terminal)Gathering DataIt was part of my duty, while working on the LearningConnections Project, to collect writing samples from thebusiness correspondents. Although I began by placingappropriately labelled boxes in the work space of allparticipants, I soon discovered that very few people weresaving their materials. On my twice weekly visits, I began bymaking the rounds, but began to stop and chat if the employeeseemed to be willing and available, whether or not there wasanything to be collected. Often my visits resulted indiscourse artifacts being produced on the spot (by printingoff a copy of something which may have oxily existedelectronically, for example), but just as often I also beganto gain insight into the jobs of the project’s participants.Sometimes I would.stay and watch them write, marvelling at theconstant interruptions of phone callers and visitors. Many ofthese putative intruders were actually contributing in somefashion to the evolution of the document under construction bythe writer, or asking for input for documents under62construction elsewhere. The collaborative way of working wasomnipresent.I soon discovered that, as time passed, I was learningabout their ways of doing writing by connecting informallywith each of them. Not only that, but I was discovering theirrelationships with each other as they were involved in thecreation of knowledge in text. I gathered these insightspartially through logs which I kept as a research assistant.However, a large part of my understanding grew as Iparticipated at Cerebellum. I gradually developed generalimpressions of Cerebellum employees, their work and theirinter-relationships. We built on a common interest: we wereall writers and more or less interested in how writing worked(and why it sometimes did not work).Ethnography seemed the perfect way to begin myinvestigation. After all, it proceeds from the belief that“the most important behaviour of individuals in groups is adynamic process of complex interactions ... [and this]behaviour is influenced by the setting in which it occurs”(Best, 1981, p. 113). Behaviour in general is a dynamicprocess, and ethnography a long term study. It seemed to meimportant to begin to understand a process such ascollaborative writing which changes, subtly or dramatically,as time passes. Given different contextual frameworks--ofplaces, people, objects--people shift their behaviour.Relationships too are built on time and careful watching,63remembering always that people are rarely, if ever, completelypredictable.Another. advantage of ethnographic methodology for myresearch purpose was that the researcher/subject relationshipis an intimate one, as Best (1981, p. 113) observes, “basedupon trust and confidence.” The ethnographer, he maintains,“gets inside the minds of the subjects, while at the same timeinterpreting the behaviour from his or her own perspective”(Best, 1981, p. 113). Eventually, I wanted to do much morethan that, but it was a valuable place to start. Trust andconfidence, in my experience, are difficult to earn and easyto lose. They build up slowly over time and can be damagedquickly. My challenge quickly became how to find an effectiveand ethical stance in research in order to achieve both “anintimate view and a cool assessment” (Geertz, 1988, p. 10),necessary to the research process.I did not begin by trying to effect a detente between twodissimilar cultures, the academy and the corporation, butmerely to try to interpret the behaviour of this “other”culture. I had signed confidentiality agreements as acondition of gaining research access at Cerebellum, but myconcerns became deeper than allowing a vital document into thehands of a business competitor: I wanted to deal fairly withpeople that I knew and liked. More often than not, seniordirectors rose to my defence when I asked to sit in onmeetings and objections were raised on confidentiality issues.I owed them at least as much consideration as I decided how64and what to gather. One research issue became how torepresent the Cerebellum collaborators in depth, but also in away which they would find accessible. Clearly, I also owed adebt to the research community to gather and present data inorder to provide new insights in writing research.To focus my studies further, I began to studyCerebellum’s employees in more detail as they engaged in thewriting activities. Project participants wrote frequently ina number of genres: memos, letters, technical troublereports, seminar notes, user documentation, and companynewsletter articles, to name but a few. They consulted witheach other over the phone, in the hallway, over coffee, inmeetings and over e-mail systems. And so I began to gatherdata: notes, rough drafts and final copies. I was, however,dissatisfied, convinced that the answers to my questions werenot in these products of writing. I began to sit in officesand take notes: what did the writer do?One early informant talked aloud as she wrote in themanner of a spontaneous think-aloud protocol. Gathering datafrom her, on her writing process, was relativelystraightforward. However, the problem of observing andrecording increased as other Cerebellum writers broke theirintense (and quiet) periods of drafting to seek advice fromothers. The conversations were rapid and fascinating. Icould not capture all of the words, nor the tone of voice inmy notes, but even less could I capture the nonverbals (bodylanguage and facial movements) nor the context of desk65decorations, or presence or absence of windows, all of whichbegan to seem more important to the writing process. Also Irealized that I had only a sense of the meaning of theconversation, but was hampered by my lack of background in thelanguage of computer professionals. In addition, I was fairlywell acquainted with the writers and advisors in more casualsettings, and it seemed to me that these personalrelationships were important in the operation of the workingprocess. I had the sense that what I could record with my penwas but an incomplete or even misleading facade of the wholereality which was writing at Cerebellum, Inc.Video-taping offered me great opportunities for the kindof detailed study needed to examine collaboration. Jacob(1987) points out that ethnographers of communication oftenuse videotaping because it “preserves data in close to theiroriginal form ... continuous ... comprehensive ... [so thatthe] naturally occurring sequence and duration of actions isrecorded” (p.20). Researchers such as Kirsch (1991) had tocount on individual writers to produce think-aloud protocols(not at all spontaneous), a technique sometimes criticized forits artificiality and tendency to cause self-fulfillingprophecy (Bizzell, 1982). I realized in doing my ownresearch, that collaboration by its very nature involves talk(and nonverbal behaviour) which can be captured on film andthen systematically analyzed and represented in ways valuableto the collaborators.66There were other ways in which video-recordingcontributed to my study. Collier and Collier (1986, pp. 169-170) suggest that we need both verbal and visual sources ofintelligence to gain more complete research insights. Iwanted to let both sources support each other. Obviously, thewritten account has a long and honourable place in thereporting of ethnography. However, the potential of thevisual, perhaps underrated in a Western culture which “wantsit in writing,” I believe, has not yet been fully explored,but some research examples show the rich possibility of visualdata.I was initially inspired by, for example, Seymour Papert(1980), in Project Headlight, who sought to examine theinfluence of computer technology on elementary schoolchildren. Goldman-Segall (1989, 1991), a member of theoriginal project team, used video techniques to show thegrowth of children’s thinking as they interacted with thecomputer programs. Her video camera recorded the nuances ofvoice and behaviour so valuable to the ethnographer while herinformants’ response to the video-camera revealed and recordedsomething about the culture of the classroom. As Hammersleyand Atkinson (1983) say in response to possible protests thatany observer may be intrusive: “How people respond to thepresence of the researcher may be as informative as how theyreact to other situations” (p. 15). It seemed to me to methat video-taping in an ethnographic fashion (that is, by67immersing myself in the environment) would offer insights intothe collaborative manners of the writers at Cerebellum.In January of 1993, I sought and received permission tovideo-record at Cerebellum in addition to taking notes andremoving written artifacts. Most of the reaction fromemployees was positive, perhaps because this is a high-technology company, and perhaps because the general managerwas an enthusiastic camcorder user, and had just been thesubject of a television program on his business approaches.Finally, the employees were used to my presence. I used afull sized recording unit with which I “shot from my hip”, asused and recommended by Goidman-Segall (personalcorrespondence, January 1992), rather than on my shoulder(“thank goodness”, one informant said, “otherwise you’d looklike the Borg!”—a reference to a race of half machine, halfhuman entities appearing in Star Trek: The Next Generation).The placement of the camera meant that I could focus on thewhole scene, rather than the smaller frame visible through theviewfinder. This wider perspective a1loved me to observe offcamera behaviour, and seek out other events to record. SinceI was not barricaded behind my camera, I was also moreavailable for interaction with my informants. In addition, bythis time I had been at Cerebellum for eighteen months andthey were accustomed to talking to me as we worked.68Sazple SelectionIn looking for a representative case of collaborativewriting, I wanted a project whicha) offered a relatively short term project (since manywent on for months)b) would be completed in-house (to avoid ethical andprivacy problems with vendors)c) did not involve expensive travel arrangements (asCerebellum deals with clients at sites all over theworld)and d) would involve a limited number of writers (to allowfor the broadest possible coverage and to look at writersworking in different subgroups).Both Borden and Sundial seemed to meet these criteria. TheBorden proposal did have some vital information gatheringevents which occurred off-site, and were thus unavailable tome. Otherwise, everything took place in-house over elevendays and involved eight collaborators. The Sundial proposalturned out to be more problematic, since it was originallyplanned as a project for one branch office, but wassubstantially moved to another office. Thus, it proved to bedifficult to follow in detail. However, the meetings I wasable to record did offer tantalizing glances (andcorroborative data) during the early, middle and late phasesof the text production process, and did involve a largelydifferent group of collaborators.69Therefore, I captured the Borden proposal writing processas a case study supported by further examples from the Sundialbid. The study examines the full range of writing activitieswhich surrounded the production of a single document. InJanuary 1993, I joined the Borden proposal writing team as aparticipant-recorder. During the following two weeks, I wasat Cerebellum from early in the morning until late at night,immersed in the process, recording every event possible whichwas undertaken in the completion of the proposal. The Sundialproposal was taped during August and September, 1993.I also talked at length with the writers, none of whomwere specialized technical writers, about their interpretationof the various activities of the collaborative writingprocess. They commented as well about other events which Ihad missed (often due to concurrent happenings). We spokeinformally as well about their reactions to being filmedalmost constantly. As might be expected, reactions variedwidely. Extroverted sales personnel loved the attention,often playing to the camera. Quieter, more introspectiveemployees, some technicians for example, agreed that theymostly tried to ignore me (or rather my camera). As the dayswent by, however, much of the office, including employees notinvolved in the proposal effort, began inquiring about myprogress as a “film maker”: they even offered advice about thewhereabouts of key players, asked about the camera, andwondered aloud who would be interested in the final result.This interest, even excitement, gave me further openings to70discuss what I was doing (for example, not making a film) andto build a relationship with more employees in the office. Ieven ate pizza late in the evening with clients and employees(as the company fed employees and guests working overtime),which afforded me a chance to interact with representatives ofother businesses important to Cerebellum’s success.Sorting DataAs I began to follow the writing activities of Cerebellumemployees more closely (for example, moving with them fromtheir desks to a meeting in the boardroom), I began todiscover that there were, however, limits to my behaviour as aresearcher who was not really an employee at all. UnlikeAlice, I found that my informants were often pleased toinclude me as participant-recorder, but leery of who elsemight share my findings: “researchers must weigh the qualityof the data they can gather (and whether they can gather anydata at all) against principles such as confidentiality,privacy, and truth-telling” (Howe and Eisenhart, 1990, p. 8).Given the competitive nature of business in the presentmarket conditions, I was often amazed that Cerebellumexecutives let me record as often as they did. Someinformants developed signals for use during filming sessions(one senior director would wave his hand if he wantedsomething kept private—a client name perhaps) and I wouldwave or nod in return. I learned that, in general, data had71to be sorted less rigorously as time elapsed betweencollection and sharing with an outside audience. Because ofconfidentiality agreements with clients and sub-contractors,information which was sensitive now would be far less so insix months. As well, a leading edge idea developed byCerebellum Cfo example, a business strategy which woulddifferentiate them from their competitors) was moreproblematic for sharing with a general audience thantechnicians deciding on a heading for a diagram.All film, notes, and written artifacts generated by me orby employees were available for my study. Except for a veryfew sensitive documents, I took data with me when I left theoffice. Sorting data for use in searching for work patternsbecame a simple matter of using everything which was clearenough (in picture and speech) to be informative. Furthersorting for focus on collaborative events was done after I hadcompleted frequent viewing of tapes and transcribedconversations which surrounded the actual drafting of text.All participants’ names can be kept anonymous in writtentranscripts, but sharing video clips is more difficult asnames are exchanged, including names of the company, on andof f camera employees, clients and subcontractors. Sortingthen becomes a matter of choosing examples which areinformative without divulging sensitive information. As withthe complete transcripts, clips must also be cleared by seniorstaff for further sharing.72My challenge in sorting was to guess what Cerebellumwould find too sensitive for general discussion and to acceptsuch limitations as a condition of my continued research. Thereality was that every employee was expected to make the samenon-disclosure agreement. Information is a valuable commodityat Cerebellum, as it is in most businesses (see also Debs,1991) : it translates almost directly to money and jobs.Information in my culture is a step on the way toepistemological pursuits; that is, a step towards makingknowledge. I could use everything, but had to be careful howI represented my conclusions to a wide audience.Analyzing DataHowe and Eisenhart (1990) maintain that “justifyingqualitative research largely consists of developing andarticulating methodological design and analysis standards” (p.2). Clearly, although the researcher can learn much fromstudying the classical rules of, for example, ethnography, “amethodology must be judged by how well it informs researchpurposes, at least as much as by how well it matches a set ofconventions” (pp. 4-5). The conventions continue to bechallenged as the research community re-invents itself inresponse to new insights and changing research purposes andavailable tools and techniques. Thus, I began to seek ways ofstudying business writers that would meet my research purposeof offering back to them a readily understandable picture oftheir collaborative activities.73Ethnography offers us one way to examine a complexprocess set in an environment of social relationships andrules, which in this case speaks to a real need:[The] relative lack of growth in contextual researchon writing can be attributed in large part to thecomposition community’s relative unfamiliarity withethnographic methodologies, which differ radicallyfrom more established experimental and cognitivelyoriented case-study approaches. (Durst, 1990, p.401)Certainly, recent calls for more ethnography (see for exampleEde and Lunsford, 1990) indicate a desire for a more detailedlook at composition activities. For researchers of languageissues who acknowledge a language’s base in culture (forexample, Vygotsky, 1978), ethnographic research seems an idealway to reveal language/culture relationships. For writingeducators, Brown and Herndl warn: “If we do not recognize andaccommodate [culturel, our teaching fights the culture--andalways loses” (1986, p. 12)It is difficult for us not to perpetuate a hegemonicseparation between cultures, whether the “other” culture isstudents or business. However, as we have seen, ethnographicapproaches help to make us more aware and perhaps more able toconfront these difficulties and overcome them. An analysis ofevents must address issues in power and power-sharing. AsDoheny-Farina (1993) reminds us, “[blecause there is work tobe done in the world, we need to walk that tightrope betweenethical self-consciousness and our attempts to observe and74analyze systematically and perceptively what is going onaround us” (p. 267).Video-recording fixes the transitory nature of events insuch a way that analysis becomes more rigorous as theresearcher views the data repeatedly, trying to find meaningamongst the “piled-up structures of inference and implication”(Geertz, 1973, p. 7) . Goldman-Segall (1989) comments thatvideotaped events lend themselves to “systematic evaluation”(p. 1) which often proves difficult in other observationalresearch. I sought to gather details on a very complex event,collaborative writing, and then use the data to build a modelwhich I hoped would organize and schematize the collaborativewriting process.I viewed the video of the proposal writing again andagain, logging events, transcribing conversations, and lookingfor patterns of interaction. I looked for ways to code data,postulating three levels of engagement with text, and divideda long, interwoven series of happenings recorded on video intochunks for transcription and further analysis. I drewdiagrams of the chain of events leading from the firststrategy meeting to the proposal leaving the office. I codedthe activities of personnel from one event to another. Duringthis time, I continued to search the professional literaturefor further insight. In so doing, I began to “constructmeaning with the data, identifying patterns and looking foranswers to ... questions” (Grant-Davie, 1992, p. 273).75In order to ensure validity, I checked my understandingof conversations with informants. I worked at length increating divisions and subdivisions for data. The data andcategories were discussed with my research supervisor. AsGrant-Davie reminds us, validity “should be judged in thecontext of a particular data base and research purpose; and itshould be demonstrated by elaborated definitions of categoriesand ample examples to illustrate them” (p. 281). My dataanalysis resulted in a model identifying three levels ofengagement with text, accompanied by detailed lists ofcharacteristics and illustrated by many sample events.Margaret Mead (1975), herself an early champion of videouse, called for “the articulate, imaginative inclusion in thewhole process of the people who are being filmed” (p. 8)Such inclusion seems to me to be particularly important in abusiness context. My Cerebellum informants were oftenuniversity trained and fully aware of “the particular types ofgames that researchers play” (Jones, 1973, p. 36), perhapsbecause they had played them! One answer to this possiblydivisive issue is to include informants as commentators.Collier and Collier (1986) speculate that “only in theory are‘we’ willing to let the ‘native’ have authoritative judgement”(p. 157). My analysis was aided by insights gained frominformants, which I recorded in note form between eventsduring the proposal writing process, and also by comments onfilm clips made by technicians. Underlying this76understanding, of course, lay insights gained from eighteenmonths of visits and more informal observations.Representing DataTelling the world, at least that portion of it that readsresearch papers, is, for me, clearly different from Alicetelling her cat. Obviously, the story of what I found atCerebellum was crucial to an understanding of another cultureby my culture. And still I wanted more: “research inprofessional writing needs to respond more strongly to needsin pedagogy, while continuing, as well, to respond strongly toneeds in theory” (Spilka, 1993, p. 215). I wanted a way torepresent my findings to the business writers themselves andto those educators who would help them to build their ownmeanings.To reiterate then, I sought to discover how collaborativewriting proceeded at Cerebellum, and to represent mydiscoveries fairly (accurately and sensitively), as “warrantedconclusions” (Howe and Eisenhart, 1990, p. 8) to interestedparties. My research led me first to involvement inethnographic approaches, then to a consideration of therhetoric of narrative and graphic representations, and finallyto the construction of a model.One powerful way to share findings is by representingthem in ways which will be familiar to all interestedaudiences. Narrative is one way to communicate discoveries.It has several advantages, for example ensuring that “the77language of the results and implications [is] in a form thatis understandable to and debatable by, various actors in aparticular setting” (Howe and Eisenhart, 1990, pp. 7-8).Language, however, is not merely a conduit for information.It is the culturally sensitive product of a community. Theproduct of research is a discourse, shaped by the rules of thecommunity which writes such documents: researchers. It isthe responsibility of researchers to alert readers torhetorical approaches. Ethnographic accounts have their ownrhetorical devices, for example, the “thick description”(Geertz, 1973), which, according to some scholars, “functionsas a textual strategy authorizing attempts at ethnographicrealism” (Herndl, 1991, p. 321) . This, I believe, is not somuch to be avoided as to be examined and admitted, and furthermethods found for representing data in ways no less rhetoricalbut perhaps more useful in some contexts, and for somepurposes, as I hope my model will be. As Kleine (1990) saidof the best ethnographic text, our concern should be that ourrepresentations are “not true but ... effective” (p.122),effective in building relationships with others in theinformant culture, and helping to improve writing practices.The model was developed by me as a way to show thepatterns of working together which developed among members ofthe collaborative proposal team as the writing projectproceeded. Both the chronological description of the proposalprocess and the diagrammatic representation of the levelsdetected in the analysis of that process are based on video78data, which is much closer to the original experience thanfieldnotes could ever be. In addition, I discovered thatfieldnotes are too limited by the speed of the writer, and bythe immediate understanding of the researcher; especially whenrecording technical conversations.The ethnographically gathered data led to a visualrepresentation of collaborative writing patterns. The modelpresents three levels of engagement with the evolving businessproposal text. Each level has different collaborativeactivities and players associated with it. Levels ofEngagement with Text offers a concise overview of my findingsand moves beyond the representations of the isolated writerwhich have, so far, dominated the writing process discussionin educational circles. As Pemberton (1993) notes: “in orderto teach writing effectively, we must know as much as possibleabout how people write; in order to know how people write, wemust observe them writing under a variety of conditions anddescribe what we observe them doing” (p. 41, emphasis mine).My descriptions construct the world of Cerebellum writers as Isaw it: in time-order at first, and then, later, indiagrammatic representation. My goal was to present aneffective interpretation of another culture for use by manyinterested parties.The model postulated is undoubtedly and intentionally areduction and abstraction that, nevertheless, renders theextremely intricate into the manageable and useful. It willrequire further corroboration before any generalizability maybe claimed. Nevertheless, these two instances (Borden andSundial) provide a beginning point. The model operates assomething of a hypothesis and, perhaps, a pedagogicalconvenience and focus for future discussion.7980CHAPTER FOUR: A MODEL OF LEVELS OF ENGAGEMENT WITH TEXT“Let’s pretend there’s a way of getting through intoit, somehow, Kitty. Let’s pretend the glass has gotall soft like gauze, so that we can get through.Why, it’s turning into a sort of mist now, Ideclare! It’ll be easy enough to get through—”(Lewis Carroll, Through the Looking Glass, and whatAlice found there)Tempting though it may be to think of any writing projectin terms of simple and straightforward stages involving prewriting, drafting, and revising, the Cerebellum proposalprocess shows vividly just how much more complex the writingprocess is, especially when it involves writers working incollaboration. As discussed in previous chapters, proposalwriters must be always cognizant of the influence of theirwriting on others, not just the client but also fellowworkers, project partners and subcontractors. In similarfashion, I seek to develop a series of insights that willultimately be useful for two audiences: academic researchersand business writers.The Business ProposalA brief explanation of the document under production maybe helpful. The particular group of collaborative writerswhich I studied at Cerebellum, Inc. had, as their ultimatetext production goal, a completed business proposal. Inresponse to a Request for Proposal (RFP), a document sent outby a company inviting response from several interested81bidders, a proposal team develops a plan to answer a businessneed. Proposals, which may vary in size according to thecomplexity of the RFP, are written by a team of employees anddelivered by a deadline time on a particular date.Although the contents of a proposal vary somewhat, anumber of topics are commonly found in such a businessdocument. These topics are discussed in numbered sections,whose order of appearance may change but, again, are fairlyconsistent. In fact, the question of topics and theirordering becomes an issue for this particular proposal teamduring the writing process. Standard sections are thecovering letter, the executive summary, the objective of theproposed project, the solution in detail, the plan forimplementing the solution, the price of the solution, and theappendices, which often include a description of the biddingcompany and its other successfully implemented projects. Theproposal tries to respond to a very basic question: Whyshould we deal with you? Formulating the answer, however, isa complex rhetorical feat.Levels of Engagement with TextFurther to my goal of communication with both theacademic and business communities, as a way of “gettingthrough” from one world to the other, I have developed a modelthat divides the proposal writing process into a series oflevels, for the sake of rendering an immensely complex processinto more readily comprehensible steps. The outline in Table821. summarizes the objective, collaborative tasks, quality andauthority associated with each of three different levels ofengagement with text.No chronology is intended by the level numbers (I, II,III) ; that is, the levels are not phases. Level I does notcome first, Level II second and Level III third. Neithershould any hierarchy of importance be attached to the order ofpresentation; that is, although level III is discussed first,it should not be thought of as more or less important thanLevel I. A helpful metaphor might be to consider the levelsas markers of depth, as one would measure water. A higherlevel (III) then is closer to the surface; a lower level (I)is deeper. Each activity within a level has its ownimportance to the task at hand, with collaborators moving upand down through the levels many times, as will be seen in therelevant descriptions below.The collaborative writing process makes obvious severalongoing themes which are common to all levels, but manifestthemselves in different ways at each level. Building andmaintaining team relationships (including being sensitive tovarious audiences), maintaining a schedule, keeping acomprehensive view of the whole project, and finally, focusingon the tasks of the level at hand all emphasize thecommunicative nature of composition for the Cerebellumwriters.83Table 1. Outline of Levels of EngagementLevel I DraftingObjective: drafting textCollaborative TasksBuilding Textdrawing (charts, diagrams)selecting (using “clip-text”)rehearsing (words, phrases)composingChecking Textreading (RFP, other documents, current text)reviewing and justifyingstyle combiningQuality: intense, focusedAuthority: skill based (How well do you present ideasin written form?)Level II Information GatheringObjective: assembling all details necessary to completeproposalCollaborative TasksBuilding Contentdiscovering sources of informationgathering information (e.g. availability, costs)Checking Contentcollating and sorting informationconsidering impact (e.g. estimating time)Quality: pragmatic, diffuseAuthority: knowledge based (How much do you know?)Level III StrategizingObjective: establishing general strategy of the proposalCollaborative TasksBuilding Strategygenerating ideasdifferentiating approachpricing solutionChecking Strategyevaluating alternatives and optionsmodifying and elaboratingQuality: extroverted, free-rangingAuthority: status based (What is your title?)84First, team work requires that time be spent on buildingconnections between collaborators (including proposal writers,subcontractors and clients) and maintaining those positivebonds once achieved. Such social networks are established invarious ways and for various reasons during the differentlevels of engagement with text. Second, Cerebellum writersmust maintain their awareness of the time pressures underwhich they are composing the proposal. The final deadlinemarks only one, though the major one, of the mileposts whichmust be attended to during the writing process. Third,members of the proposal team at every level often allude toother levels, and to other events on the proposal writingtimeline, in an attempt perhaps to maintain a comprehensiveview of the entire process. Finally, collaborators must learnto focus on writing tasks at hand, whether those tasks bestrategizing, information gathering or drafting. I willillustrate these on-going themes of collaborative writing indetail with examples from data during the discussion of theparticular proposal writing events observed.Each level is defined and described in general termsbelow. I have decided to begin with Level III, strategizing,because the collaborative writing processes I have observedbegan with discussions of the general business approach.Level II will follow, as information gathering beginsimmediately after a strategy is established. Level I,drafting, starts last, and will be discussed in the ultimateposition.85Level III—--StrategizingLevel III is concerned with establishing and maintaininga general strategy which will be realized in the businessproposal. Such strategizing involves senior employees indeciding on a winning approach, both in content and inpresentation. The primary goal, then, is to establish whatbusiness people themselves call “high-level” ideas; that isthe general, non-detailed tenets of business strategy for thewinning of job contracts. The customer should be impressedwith the company’s understanding of the business problem,their potential competence in the effecting of a solution, andtheir competitive business stance, that is, price. Becausepricing by various competitors on any single bid is now verysimilar, today’s customers want to see extra quality inoffers, highlighted by the business strategy. The notion ofstrategy embraces not only business approaches, which partnerto choose, for example, but also rhetorical ones such asquestions of diction. How to present ideas is an importantpart of the strategy of winning contracts.Several activities are undertaken at Level III whichserve to build the business and rhetorical strategy. Membersof the team bringto Level III meetings stories of priordealings with clients and potential customers. They discussperspectives of how Cerebellum’s overall goals might affectthe current proposal. In close concert, the team struggles togenerate ideas, differentiate their approach from that ofpotential competitors, and reach an appealing approach to86pricing. In discussions aimed at discovering the beststrategy, Level III collaborators aim for consensus but areprepared to deal with conflict to achieve the best answer.And as they do, the approach, which must gradually become thetext of the proposal, is reviewed or checked. It isevaluated, modified and elaborated. The proposal grows andchanges in collaboration with others, both present as areCerebellum team members, and imagined, as clients must oftenbe.A further checking goal of Level III is to ensure thatthe proposal once drafted adheres to the plans agreed upon,and, further, that the text presents the strategy persuasivelyto the client. Senior management may also review theeffectiveness of the proposal writing process. Severalactivities dominate the attaining of these checking goals:re-reading the Request for Proposal (RFP) and comparing itsrequirements with drafts of the text, talking to other teammembers and confronting problems of approach or rolemaintenance with them, and changing the text for persuasiveimpact.The quality or tone of Level III is closely linked to thebasis of authority for Level III activities. Meetings whichdeal with business and rhetorical strategy are usuallyextroverted and free-ranging. Members encourage each other toexpress their ideas without fear of excessive evaluation. Thegoal, obviously, is to generate many ideas in hopes that oneor more will be worthy of further consideration as a general87strategy. Senior employees are generally well experienced andhave much to offer a brainstorming session. Authority atLevel III, however, is based on company status. AtCerebellum, titles such as “Director” are invested with a gooddeal of authority. The persons holding these roles are morepowerful in group situations. Differences of opinion andfinal closure are still regulated by the most seniorexecutives, usually those with the most experience. It shouldbe noted, however, that officially, Cerebellum claims not tohave a hierarchical structure.Ongoing ThemesWhile each level has unique objectives and activities,some themes percolate throughout all the levels yet aremanifested in different ways at each level. The ongoingthemes are establishing and maintaining relationships(including being sensitive to the various audiences),maintaining a schedule, keeping a comprehensive view of thewhole project, and focusing on the tasks of the level at hand.Establishing and maintaining relationships. Level IIImeetings attempt to build common purpose between and amongteam members of more senior status (for example, directors andmanagers). The team further tries to identify itself as asubgroup of the company by searching for “differentiators”(ways of separating themselves from competitors). Commonground is established in various ways, by relating shared pastevents, for example. Camaraderie is also built by suchbehaviours as responding empathetically during the meeting to88the emotions of other team members (such as outrage andhumour) and responding positively to their ideas. Theseactivities are especially important when the meeting includesnew employees, who are as anxious to blend in to companyculture and establish themselves as valuable team members asthe company is for them to identify with its goals. Care isusually taken in subtle ways to forge bonds among teammembers, and between the team and the company culture as awhole. However, team roles can also be attacked in moredamaging ways. Finally, however, each member of the Level IIImeeting wants to establish, or even re-establish, theirexpertise and power within the group. Individual membersoffer caution and advice to each other in a more or lessfriendly manner. They tell anecdotes, and even roleplaydialogue, to illustrate their points. At times they openlychallenge other opinions aggressively, or show disagreement inmore subtle ways, for example by engaging in private one-to-one conversations while the main meeting proceeds.Audience sensitivity. At times, Cerebellum writersconsidered and treated the client audience as importantmembers of the team in the proposal effort; at other timesthey placed them in more antagonistic roles. Participants inLevel III strategy meetings also struggled constantly toestablish relationships with business partners. Partners arevendors who sell, perhaps, computer hardware or some otherproduct or service which Cerebellum, as a systems integrator,will need in order to provide a complete solution to a client.89Partners are yet another audience whose relationship withCerebellum often proved difficult to establish.Schedule maintenance. All collaborators working onproposal documents must balance their need to plan andinvestigate solutions thoroughly with their need to pressforward to meet the proposal deadline. Sometimes sacrificesare made, and the need for care and attention loses out to theneed for alacrity. Such scenarios may later be costly,however, and there are constant attempts to avoid panic andpoor decisions——rushing to closure, for example, may beat thedeadline but lose the profit margin. To this end, teammembers must also bear in mind the interlocking nature oftheir contributions to the proposal. At Level III, however,schedule maintenance is less important, perhaps because moststrategy meetings occur relatively early in the process beforetime pressures become a serious problem. Keeping to aschedule, it seems, only becomes important late in theproposal production during Level III events.Keeping a comprehensive proposal view. Level IIImeetings must often consider borderline sales which may beaccomplished by the careful presentation of ideas by proposalwriters. Such drafting would be accomplished during Level I.In addition, although senior management brings experience tothe table, they also bear in mind the necessity of carefulinformation gathering (Level II) to achieve the detail neededto implement the general strategy and reflect the necessarycompetence to win confidence. Level III meetings consider90also personnel available to accomplish all tasks associatedwith the proposal writing task. Pointing forward to remindeach other of upcoming tasks remains an important task ofdirectors at Level III events.Maintaining level of engagement with text. The generallysenior status of employees who take part at Level III and theegalitarian quality of their meetings mitigates against strongenforcement of attention to task. Lapses must be delicatelyhandled to avoid the appearance of threat to power, or to thewide-ranging scope of Level III meetings.Level 11—Information GatheringAlthough members of the strategic team bring broadknowledge to Level III, once a general strategy has beenagreed upon, details must be sought which are specific to thecurrent situation. Such work involves primarily technicalstaff, but managers and more senior staff are sometimesinvolved as well. The primary goal of Level II is theassembling of information needed to complete the proposal.The proposal team reaches out into the company, even beyondthe local branch, and finally outside company walls, workingwith people increasingly less familiar to them. What isavailable? What will it cost? Expertise, software andhardware can be bought and sold, but first, someone has todiscover and access relevant sources of information. The. lessovert goal of Level II is the control of knowledge flow.Asking questions reveals interest, even business or technical91approach, to a skilled interlocutor. Answers may come with“strings attached,” and favours owed.Several activities build content at Level II. WhereasLevel III writing focused primarily on recording generalconcepts which arise as the result of brainstorming sessions,Level II writing concentrates on discovering sources ofinformation, accessing those sources and then gatheringdetails, filling in (“populating”) charts and taking notes,for example. These details must then be considered, what Ihave called “checked,” at meetings where collaborators discussthe impact of the information on the proposal. Level IIinformation gathering and consideration may influence thegeneral strategy shaped at Level III, and will certainlyaffect drafting activities at Level I.Though information gathering has, certainly, someexciting elements of the “chase,” knowledge seekers areusually involved in considering the pragmatic ramifications ofinformation. Level II collaborations are often conducted in amarket of buying and selling, dealing witth a continuum of moreor less known, and more or less friendly parties.Collaborators building a store of facts and ideas rely oftenon relationships developed over a long period of time,remembering also debts owed and caches of good faith availablefor use. Checking sessions are conducted with collaboratorscloser to the centre of the proposal effort to look at theramifications of such information as availability and cost ofgoods and services. Level II is then more diffuse than Level92III, with collaborators ranging far afield to gatherinformation. Such is the case in particular when dealing withadvanced technology where products come and go with oftenstunning rapidity. Authority at Level II is knowledge based.How much someone knows will decide who controls thecollaboration. The status imbued by the company hierarchy isthus largely dissolved, since new ideas come often from junioremployees with recent education.Ongoing ThemesP1rnrr CT Level IIactivities which centre on information gathering alsohighlight relationship building and maintenance. Whereassenior management in Level III meetings often have the luxuryof long-term knowledge of each other, information gatheringoperations must frequently be conducted with others relativelyunknown. The seekers are required, if they are to besuccessful, to make the source feel sympathetic to team goals.If the source is a company member of another branch, supportmay be gained easily by, for example, appeals to expertise.If, however, the source is a possible competitor, or apossible, but rejected partner, inducing co-operation is moredifficult. Requests must be more tenuous, careful not topromise more than the seeker can deliver. Suggestions of pastfavours, or future relationships tend to mark thesediscussions. And if there is any doubt that knowledge in thecollaborative setting is power, consider the possible falloutof a single phone call seeking pricing information—what has93been revealed by the information seeker? And how will theinformant respond to the request? Something valuable may beinadvertently revealed: sources may be insulted. Informationseekers must know when to be diplomatic, even circumspect.Conversations with collaborators more distant from the coreteam of company employees (such as clients and partners) mayprovide a further means of cementing the relationships of theimmediate team, providing as they do information not just onfacts but on emotional reactions which may later be importantto the team’s success.Audience sensitivity. As discussed above, there are manyaudiences, more and less distant, at Level II. Telephonecalls and face to face meetings prove to be more like playingchess than simply acquiring facts. As team members at LevelII, writers face considerable challenges, including vendorsand other audiences, as they attempt to gather informationwithout giving away proprietary facts.Schedule maintenance. A constant struggle between theneed to be accurate and thorough, and the need to make hastepermeates Level II activities. Although senior executiveswould prefer to do data collection first and then proceed todrafting, technicians must often do tasks at the same time.They remind one another of how time is passing, especiallyduring checking meetings where one criterion becomes theamount of time a procedure will take. Because proposals arewritten under considerable time pressure, better ideas which94take longer will often be rejected in favour of adequate butquicker solutions.Keeping a comprehensive proposal view. Being “in themiddle,” writers seeking information at Level II are acutelyaware of the duties given them by Level III strategizing, andalso by the concern for written presentation of ideas in LevelI. They may, for example, begin to sense that the generalplans of Level III are unrealistic as they discover newinformation. Or they may find questions of presentationanswered in pieces of text available in other documents, readyfor re-use in the present proposal. The gathering of factsand opinions, textual formats and even whole sections of textseems to be pursued in much the same fashion.Maintaining level of engagement with text. Remainingengaged with Level II tasks is fairly easy during the buildingphase where collaborations are quickly established, questionsand answers are exchanged, and collaborations are broken off.Checking meetings are more problematic, given thatparticipants must face the implications of information and aretherefore often tempted to discuss the strategy (which shouldhappen at a Level III meeting). Collaborators may be temptedto jump to levels which are associated with more powerfulemployees.95Level I—DraftingLevel I has as its primary goal the production of writtentext. Existing files become like data bases which can beaccessed for cut and paste activities to help to save time andtarget audiences. Writers must assemble information gatheredat Level II, keep in mind general directions established atLevel III, and work to draft a detailed solution to thebusiness problem even as they look for the best method ofrhetorical presentation. It is during Level I that thesecondary goal, evaluating the solution, also becomesimportant. General ideas may turn out to be unworkable oncedrafted in detail. Vital steps may be missing, requiring moretime and money to accomplish. Writers must remain aware ofthe need for more information, or for the alteration of high-level strategy. Collaboration builds not only the text butthe solution itself concomitantly.Drafting proposal text still proceeds through buildingand checking activities. Building text, which technicians dowhile developing the solution, typically involves severalactivities such as the drawing of charts and diagrams, eitherfor inclusion in the final text or for the purpose of sharingideas; selecting and modifying previously written text for usein the document currently under production; and composingoriginal text which involves the oral rehearsal of words andphrases. Checking activities include reading documents (suchas the RFP and the text under production), reviewing and96justifying the presentation of ideas, and merging the variouswriting styles into a unified corporate voice.Level I, drafting, especially during building of text isvery intense and focused, what business writers at Cerebellumcall “heads down” work. There is an urgency to complete thewriting task. Authority at Level I is based on skill withwords. How well a writer presents ideas often determines whowill control a Level I collaboration. However, sharedleadership sometimes occurs as well and high company statusoften intrudes in the logical granting of power to the mostskilled, as we see also at Level II.Ongoing Themesn-inri r1 Level Icollaborations of writers actually drafting text are marked byfairly intimate relationships. Writers learn to lead andfollow, clarify and detail, order and re-order in a kind ofchoreography which grows easier as the dancers become morefamiliar with, and accepting of, each other’s ways of writing.Ideally, they must find ways to produce documents, agreeing onoutlines, drafting workplans in diagrams, dividing data entryduties, maintaining give and take on final decisions regardingwording. Writers must also discuss how to defend theirdecisions, especially as the writing of the document does notfollow the discussion of the business solution, but rathershapes general premises into operational plans. Draftersconstantly evaluate their document based on discussions witheach other. Since all text might be built and checked by97different writers or groups of writers, relationships alsotend to make drafters more equal. A junior engineer, forexample, may be asked to review the work of a more seniortechnician and will be expected to be critical.Audience sensitivity. Level I involves collaboratorsstruggling constantly with questions of audience.Collaborators must be aware of all the audiences for theirdocument--clients, partners, subcontractors--whose desiresmust also be kept in balance by the company directors andmanagers (yet another audience!) as they review the document.Indeed, varying interpretations of the desires of the mostimportant ultimate readers, the clients, can consume a greatdeal of time. While technicians often prefer to write to anaudience of fellow engineers, they too must deal with the needto write, at least in some parts of the proposal, for anaudience of senior executives in another company. The styleappropriate for different audiences varies, and this canbecome problematic.Schedule maintenance. As one director at Cerebellum washeard to comment: “We are eleventh hour people.” Despite theefforts of collaborators to hurry along (and enjoin other teammembers to hastenas well), drafting tends to fill timeavailable. The ultimate deadline, the time at which theproposal must arrive on the client’s desk, serves as the finalmaintainer of the schedule during Level I. Other commentsintended to enforce team discipline and to help the group stayon time are made directly by senior employees.98Keeping a comprehensive proposal view. Level I draftingis evaluated by its adherence to ideas advanced as generalrhetorical strategies during Level III meetings. Level Idrafters must also be aware of the need to cycle back to LevelII activities if they need more information to make furtherdrafting possible, or to Level III if more general problemsarise. Level I also looks forward to the possibility of anoral presentation stage if the written proposal isshortlisted. At such a presentation, managers and seniortechnicians will be asked questions of clarification, whichcan be anticipated by writers who discuss their ideas fullywith each other, and especially with those likely to bequestioned.Maintaining level of engagement with text. The intense,focused quality of Level I activities, motivated largely bytime pressures, tends to ensure that writers will stay ontask. If they are tempted to jump to another level, it willbe because what they are writing puts them in a quandary ofsome sort: perhaps it reveals a difficulty in the overallapproach, or some missing information.ConclusionThe preceding overview of the Levels of Engagement withText will, I trust, assist readers to understand the generaloutline of my model. Three levels are posited. Level III isconcerned with strategy, Level II with information, and LevelI with drafting. The three chapters which follow describe and99discuss twenty-three events, featuring business writerscollaborating on the production of a proposal document. Theseevents are presented chronologically. Readers are directed toFigure 1, which shows the Path of Document Movement BetweenLevels, and Table 2. which offers a Classification of Eventsby Level, Day and Participant. Participants are listed inorder of rank in the unofficial company hierarchy.These additional perspectives provide overviews offindings which should prove useful in following the narrativeof chapters five to seven. The final two chapters, then, willsummarize findings at each level and look at variousimplications of the model for business, education and futureresearch.Figure 1. Path of Document Movement Between LevelsIII.. 4 . 4Levii • 444 4 4 • 4eI- 4•4 4 44• 4 4 44A B C DE F GHJ K L M NPQ RST UVWYZEventTable 2. Classification of Events by Level, Day, andParticipant (in order of rank).ventE Aar B1y CDEFGM Hid Ja1 KeLMNPLa Qte RSTUVWYz100jJV.L yThe Starting Gun III 1 Di 1)2 1)3 Al MlOpening Moves II 1 Al Ml TiChanging Horses III 4 D2 Al Mi A2Who’s On First? II 4 Ml TiDecisive Action II 4 A2 MiUnder Consideration II 5 D2 Al Mi Ti T2Quick Change Artist III 5 GM Dl D2 D3 Al A2 MlProblems, Problems I 6 Al Ml Ti T2Down to Work I 6 Ti T2Covering Off I 7 Al TiThe Audience Awaits II 7 Al Mi TlA Decisive Dyad I 7 Al MiCoiiabo-tech II 8 Ti T3Enter the Dragons I 8 D2 Mi Ti T2The Recyclers I 9 Ti T2A Woman’s Place I 9 D2 Ti T2Pulling Rabbbits II 9 D2 Mi Ti T2Out of HatsAsking the Tough III 10 Di D2 A2 MiQuestions IRounding Up The II 10 D2 A2 TiStragglersWords, Words, Words I 10 D2 T2If Only III 10 D2 A2Coming Together I 10 Di D2 MiDeadline Day I, II, ii Di D2 Al A2 Mi Ti T2& III101CKAPTER FIVE: PRESENTATION AND DISCUSSION OF EARLY EVENTSThe White Rabbit put on his spectacles. “Whereshall I begin, please your Majesty?” he asked.“Begin at the beginning,” the King said, verygravely, “and go on till you come to the end: thenstop.” (Lewis Carroll, Alice’s Adventures inWonderland)With the general characteristics of the Levels in mindthen, let us begin at the beginning and examine illustrationsof the model. These are revealed in data collected during thecollaborative writing of two proposals which I have called“Borden” and “Sundial”. The following chapters, dividedaccording to early, middle and late events, report myobservations and analysis of the collaborative proposalwriting process at Cerebellum Inc. primarily during thedevelopment of a proposal to Borden Bodyworks. My discussionis augmented by several events from the writing of a proposalfor Sundial Industries.In each chapter, I propose to take the reader through aseries of events detailing a particular process ofcollaboration leading to text production. Each “event” is ameeting of a group of collaborators. For example, Event A inthe Borden proposal is a group of senior executives discussingstrategy. At the end of this gathering, two members convene anew meeting, thus demarcating a new event. Event B, then,shows the proposal manager and an account executive gatheringinformation about the proposal process. A technician is102eventually involved and becomes part of the originalconversation; therefore, Event B is sustained. Events canusually be placed at one of the three posited levels ofengagement. Readers should note that I have not labelled anyevents as letters I, 0 or X in order to avoid confusion withnumbers, especially since the levels are called I, II and III.The Borden ProposalDra.matis PersonaeThe Borden proposal was undertaken primarily by a groupof eight Cerebellum employees (and a significant group ofsupporting people who had very minor roles and are brieflymentioned as they appear). The major players and their chiefroles, as defined by the company, follow. The order, asgiven, is hierarchical; that is, directors have more powerthan account executives. Within groups, seniority (years withthe company) dictates authority. Director 1 is thus the mostinfluential director. Readers should note, however, thattheir actual roles shift as the proposal takes shape, so ajunior member (in experience and title) such as Technician 2may control a situation while collaborating with a more seniormember. The main players, then, are as follows:Director 1 (D 1) supervises the development of allstrategy to build and maintain the business ofCerebellum as a computer systems integrator.Director 2 (D 2) oversees Cerebellum’s technicalapproach to solving client business problems.103Director 3 (D 3) co-ordinates the selling ofCerebellum’s expertise and maintains itsstrategic partnerships.Accountexec 1 and 2 (AE 1 and 2) are ultimatelyresponsible for the proposal in every detail:pricing, viability, final success. They bringin the business initially and maintain customerrelations during a proposal and the followingproject to promote further businessopportunities.Manager 1 (M 1) supervises the proposal developmentprocess to ensure on time completion, andserves as lead writer and co-ordinator of theproposal writing team.Technician 1 (T 1) creates the technical solutionand is often called the technical architect.Technician 2 (T 2) serves as assistant to thetechnical architect.General BackgroundAfter Cerebellum had already been working on a project atBorden Bodyworks for several months, Borden decided to proceedwith a completely computerized approach to record-keepingwhich would connect their many branch offices both with eachother and with a constant stream of clients. They hired aproject manager (“Jake”) to represent them, and issued an RFPto find the computer hardware which would solve their problem.Amazingly, large portions of the RFP had actually been copied(errors and all) from documents outlining technical needswhich had previously been written for Borden by Cerebellum104employees. Now, however, Borden wanted a computer systemwhich would provide a complete business solution, and whichwould, of course, be guaranteed against “downtime”—periodsduring which the system would not be operational. Toaccomplish such a project, Cerebellum, as a computer systemsintegrator, would need to partner with a company which wouldprovide the hardware.The Sundial ProposalDramatis PersonaeDirector 2 (0 2) directs all technical operations(see above, Borden proposal)Director 3 (D 3) directs sales operations (seeabove)Director 4 (D 4) directs systems integration (seeabove, Director 1)Director S (0 5) manages another branch office whichhas initiated the Sundial proposal effort.Accountexec 3 (AE 3) takes ultimate responsibilityfor proposal effort (see above, Accountexecs 1and 2).Manager 2 CM 2) manages the proposal writing process(see above, Manager 1).Technician 4 CT 4) creates the technical solution,the technical architect (see above, Technician1)Technician 5 (T 5) assists with the creation of thetechnical solution as a technical specialist.105General BackgroundAlthough I have concentrated on analyzing the Bordenproposal, I offer here a supplement and support for mycomments with data from another project, Sundial, a proposalwhich was written by Cerebellum employees some seven monthsafter Borden was completed. Sundial wanted to convert clientrecords from one format to another for better storage. Thecompany once again had a single contact person who was,however, a long-time employee of Sundial. With the exceptionof Directors 2 and 3, the proposal team for Sundial wascomposed of different members than the Borden group. Theirroles are very similar, though each individual, of course,plays their part differently.The Sundial proposal group involves two of the sameplayers as the Borden team, Directors 2 and 3. Their initialmeeting shows evidence of many characteristics of Level IIImeetings during the Borden proposal development, realized, ofcourse, in ways particular to two different groups of seniorexecutives. As with all Level III meetings, the Sundialencounters were characterized by emotional volatility:confident predictions and warnings of disaster, a touch ofironic humour, and the tension of challenges to each other’spower.The formal writing of each of the Borden and Sundialproposals began at meetings which included three directors andan account executive. The Borden meeting also included aproposal manager, but the manager for the Sundial proposal was106not yet working at Cerebellum although she had been hired. Icollected data at this particular Borden meeting with fieldnotes, as the Accountexec had indicated that he was tooconcerned about client confidentiality to allow video-taping.The Sundial meetings were all video-taped. As mentionedabove, Level III collaborations are undertaken by the mostsenior personnel in relatively large groups (usually five ormore), and these meetings follow this pattern. Seniorpersonnel are experienced, and used to wielding a considerableamount of power. However, in both of these proposals, themanager who was specifically charged with overseeing andwriting large portions of the proposal was a new employee.These proposal managers were unfamiliar both with the companyand its ways of working and with the other members of the teamat every level. Collaboration is often hampered by such lackof knowledge. Senior management otherwise tends to avoidinitial drafting, preferring to critique and if necessary,revise.In an initial Level III meeting, my data shows, senioremployees come together as relative equals, yielding only intoken ways to a chair, who may attempt to direct theconversation. Their primary objective is to deal with highlevel issues for the writing of the proposal, such as theestablishment of general business and rhetorical strategies.For these reasons perhaps—size of group, equal power, and animportant goal—Level III collaborations are emotionallyvolatile, marked by idealism, acrimony, challenges,107roleplaying and storytelling, predictions and warnings, jokingand laughter, tension and high energy. Though charged withthe quest to create themselves as the company team, theparticipants must balance the drama of the search with morepragmatic needs. Finally, then, Level III meetings are markedby what I have posited as three of the ongoing themes ofcollaborative proposal writing at Cerebellum: establishingand maintaining relationships, being sensitive to audience,and keeping a comprehensive proposal view.Each of the eleven proposal production days features oneor more events at different levels of engagement. An overviewof the movement from level to level follows (see also Path ofDocument Movement Between Levels, Fig. 1) . Every day isintroduced with a summary of the proposal state achievedduring the day. Then details of the day’s events follow.Day OneThe first day of the Borden proposal results in theproduction of notes on the strategy decisions made at theLevel III meeting. Collaborators also begin to assemblemodels of other proposals and outline a schedule of eventswhich they hope will be followed during the next eleven days.Manager 1 and Accountexec 1 have also sketched out therelationships between Cerebellum and Borden personnel.108Event A. Level III. The Starting GunEvent A is a meeting to decide whether or not the Bordenbusiness will be pursued and if so, what the general strategywill be to win the contract. The meetings involve arelatively large group, five of the most senior executives.The three directors would be considered at the top of theoffice hierarchy (except for the general manager) andtherefore above account executives and managers. Director 1is definitely the most powerful of the three directors. Theaccount executive has company experience and has brought inthe Borden business, Manager 1 is a new hire and unproven.The clash over the relative power of the players becomesobvious as the meeting progresses. Even this struggle servesa purpose, however, as it establishes the right of each playerto a place at the table. Peace would perhaps showcomplacency.The primary objective of any level III meeting is high-level strategizing. On the surface, the Borden meeting issuccessful and accomplishes its main objective. During thisevent, when Director 1 says “Let’s bid Sauct equipment. We areprime,” the meeting quickly ends: a general decision has beenreached. The Sundial meeting shows even more clearly thelevel III general strategy objective. For example, Director 3takes control of the collaboration and begins by setting anagenda for what they might work on together at the initialmeeting:109D 3 What are the objectives you want to get out ofthis today?AE 3 I want to get at least a high-level agreementon strategy or input on strategy where we can’treach final conclusion. So strategy, I’d liketo have staffing for the bid and proposaleffort and I’d like some concurrence on thebudget we’re working towards for theseThe initial proposal meetings seem to have the same generalobjective. However, the Sundial meeting already shows signsof an account executive who has other objectives as well: alist of people to put the proposal together, and an agreementon the costs of the proposal effort.As befits a meeting of high-level discussion, theseexecutives spend some time sharing their visions with oneanother. Director 2 shows irrepressible optimism during bothmeetings. What lies ahead for the team? Director 2 says toDirector 1 in an aside during the Borden meeting: “We canmake a shit-load of money!” Business proposals are, forDirector 2, at least, clearly a time to dream about wonderfulopportunities. Even when Director 3 sees a possible “redflag” or risky situation such as an unconsidered problem whichmight deserve more discussion or at least a warning, Director2 still maintains his excitement:D 3 Will the customer live with these parameters?D 2 We’ll shoot for the moon and back up!The opening Sundial meeting shows the same enthusiasm. Herewe see Accountexec 3, who interrupts his own dutiful and dulllisting of staffing requirements with a burst of enthusiasm110addressed to Director 2. Discipline, always tenuous at LevelIII meetings anyway, is subordinated to idea generation, tohearing out possibilities which will not be realized untilLevel I but can be shared at the present time:AE 3 That’s the idea. When I say staffing, I wantto fill in the gaps. They’ve got a great logoby the way.D 2 Who?AE 3 Sundial. You’ll love it for graphics. It’s inflaming orange...Although Director 3 worries about the client’s vision notbeing met, above, Director 2 is able to reassure him.However, this too serves to keep on the table the task athand. Every business opportunity bears some risk as well.Director 2 and Accountexec 1, for example, wonder about acompetitor’s plans during the Borden meeting:D 2 Is Azure going to propose on their own?AE 1 They’ll decide tomorrow. We don’t want to bidwith them. There are problems.and Director 3 mutters darkly about yet another competitor:“[Sauct] has a lot to lose if they can’t deliver.” Level IIImeetings, especially those which initiate proposals, mustweigh problems carefully. Manager 1, the new hire who wouldbe doing much of the drafting on the Borden proposal, actuallymakes his first comment of the meeting as a warning. Heraises what the group judges to be a valid red flag. The“somebody” referred to is likely the account executive.However, he makes no note of this advice (which does indeed,sound more like a dictate from a superior than a suggestion in111the spirit of the collaborative proposal writing they arepresumably working towards):M 1 I’m reading this [the RFP] and if I was [Jake][Borden’s project manager], I’d be nervousabout [Jobber] [a company presently working forBorden].D 3 You mean their maintenance [of the new system]?D 1 (to AE 1) Tell us about [Jake]AE 1 He’s very low risk.D 3 So somebody’s got to address this.Working productively with Jake continues to be a challenge tothe team throughout the proposal writing period. Director 3has here identified a concern which will be carried forward bypersonnel into every phase of the project.Conflict at a Level III meeting can be handled in variousways. During the opening level III meeting on the Sundialproposal, Director 5 and Accountexec 3 disagree on theclient’s RFP demand for a business case. Together, theysearch through the RFP document, looking for the relevantpassage. Accountexec 3’s body language during the followingconversation is of interest. He does not make eye contactwith Director 5, even though Director 5 comes over to look atthe RFP document with him. Accountexec 3 continues to makeplans for getting a business analyst (looking to Director 3for a response) even while he looks for the passage underdispute. Accountexec 3 seems irritated with being questioned,even by Director 5 with whom he frequently worked quiteamiably:112AE 3 It’s three bullets ... (flipping and lookingthrough RFP, continuing to talk). So we needto get a business analyst working through this-- here it is (reads aloud, then-) That’s mealymouth wording for “do it”! It came up in thebidder’s conference as bad [ambiguous] as this,but I know from [Steve] [his Sundial contact]that only the bidders that do it [build abusiness case] will have a chance of winningif you’re not smart enough to go to theclient and say “what did you mean by this?”...In this case, the collaborative spirit seems to bemissing. The question is valid; the questioner, sincere.Director 5 does not press the matter further, unwillingperhaps to risk their working relationship over a minor issue.Clearly, Accountexec 3 collaborates well with the client,which seems to be enough at this juncture.Disagreement also occurs during the opening level IIImeeting of the Borden proposal. Director 3 arrives late.Accountexec 1 has taken charge of the meeting, but Director 3interrupts his comments frequently. Director 3 then begins tohold whispered conversations with Director 2 while Accountexec1 tries to continue chairing the meeting. Matters come tocritical point when members find that reaching a generalstrategy decision involves consideration of a number ofpossible partners for Cerebellum. Several problems becomeapparent. First, the collaborators are suffering from optionexhaustion--too many choices of possible partners and thepower relationships Cerebellum might exercise with each one.Second, rather than working together, Director 3 continues tochallenge Accountexec 1, actions difficult to interpret as113challenge Accountexec 1, actions difficult to interpret asanything but indicative of scorn for Accountexec 1 as chair,as chief contact with Sundial, even as valuable collaborator.Director 3 is pushing for a confrontation, which Accountexec 1rejects. Third, Accountexec 1 turns the fbcus of the meetingon the process of consensus rather than addressing the problemin some more fruitful way:D 1 What do we have to do to decide?AE 1 There’s still one more option.D 2 (laughs) Absolutely.AE 1 (names another possible partner)D 3 Let’s vote.AE 1 No. We work on consensus.Meeting members do not vote, but add one more option and thenpush for a perhaps too hasty conclusion on strategy.Director 2’s reaction is noteworthy as well. There iscertainly acrimony in this situation, but Director 2 alsovoices the ironic humour of a making a decision on a bid whichalready has too many options when Accountexec 1 wants tointroduce yet one more. His laughter, however, is the onelighter moment in the meeting. The humour seems to serve as agentle disciplinary act, warning a collaborator that they havetransgressed in some fashion: in this case, made animpossible or ridiculous demand on others.The opening Level III meeting of the Sundial proposal hasmore obviously humorous moments which function in the sameway. Conversations which could turn into confrontations can114be changed with humour, depending perhaps on the willingnessof other team members to play along. For example, hereAccountexec 3 is concerned about meeting a client’s needs:AE 3 They’re concerned about security.D 2 (laughs) We’ll use blind people [to work withclient’s data]AE 3 (laughs) I’ll put that in the proposal!Still, an audience concern has been raised by the accountexecutive and the collaborators made aware of the client’sneeds.Humour, however, needs to be carefully handled lest itbecome a cause of dissent. For example, the Sundial meetingengaged in some ribald humour at Director 4’s expense.Director 4 is a new company director. He has joined thismeeting late. Unlike Director 3’s lateness discussed above,which may indicate a lack of regard for Accountexec 1,Director 4’s tardiness stems from time spent negotiating amajor business deal for Cerebellum. His news energizes theSundial meeting. They begin to discuss a possible teammember, Technician 5, and wonder about his objectivity inevaluating hardware. Here Director 2 sees the possibility foran amusing double entendre, which is immediately picked up bythe other team members. However, when Director 4 resists thehilarity, Director 2 backs off:D 4 But he can go both ways (defend two differentproducts), it sounds like.D 2 (smiles) He can go both ways.(general laughter)115D 4 (earnestly) But we want somebody who can goboth ways!AE 3 Speak for yourself! (laughs)D 4 Because that way we can see the other side.The best debaters can go both ways, right?D 2 (seriously) He can go both ways.Director 2 avoids provoking an unpleasant situation. It isespecially interesting that he uses the same words, “he can goboth ways”: first to begin the joke and second to end it.On-going themes of collaborative groups are also presentin these initial level III meetings. Building and maintainingteam relationships and keeping a comprehensive awareness ofthe entire proposal process are evident at both the Borden andSundial meetings. Although working on collaborativerelationships is important also at Levels II and I, at LevelIII, Cerebellum employees work most strongly to establishthemselves as a team. Level III offers, in the early stagesof the proposal writing process, a first and crucial step infixing the ground rules of the team. To this end, Level IIIrelationships are constructed, in part, by members askingquestions as the group tries to clarify its job. Suchrelationship building is especially evident at openingmeetings, partly because the employees of Cerebellum come andgo fairly rapidly, so team membership changes. Therefore,employees are often plunged into collaboration without ashared work or personal history. One way overt attempts tobuild the social network become most apparent is in the searchfor “differentiators” (what makes the company team different116from other teams). Team identity and team advantage followfrom the discovery of differences. The Borden proposal teamstruggles with this question at their initial meeting:D 3 Does [Jobber] have a maintenance offering?D 2 Could be a differentiator for us.and later:D 2 Do we have inventory list?AE 1 Yes.D 2 And others don’t have it?AE1No.D 1 A key differentiator?As I have previously noted, because price is often much thesame in today’s very competitive market, extra quality counts.So, apparently, does separating the proposal team from theother competitors, at least for now. In the future, acompetitor may become a partner in a business proposal and theteam will have to be created again.Differences may be more subtly established by teammembers working to invoke a certain ethos, or moral basis thatwill also inform choices in the proposal. During the initialmeeting on the Borden proposal, Accountexec 1 relates a storyabout three vendors under discussion as possible competitorsand as possible partners. Vendor 1 had recommended Vendor 2for a job. Later Vendor 2 recommended Vendor 3 to a client,rather than recommending Vendor 1 in return for the earlierfavour. The consensus of all at the meeting was that suchbehaviour by Vendor 2 was “sleazy” and “nasty”. The story has117the flavour and effect of a morality tale. Team membersdistance themselves from the behaviour of Vendor 2, but theyalso draw together as a team by sharing a common reaction. Itshould also be noted that the story has the effect of awarning. If Cerebellum decides to work with Vendor 2, theymust be aware of Vendor 2’s business practices.Humorous interludes, like stories, also have a positiveeffect on collaboration. Sharing a joke energizes themeeting, dissipates tensions and unites the group. Also, aswe have seen before, humour tends to regulate what isappropriate. In the following interchange, we see the Sundialproposal group, during their initial level III meeting, takethe unusual (and4probably ill-considered) step of engaging ina Level II activity, detailed information gathering. Suchlevel jumping might impede the collaborative process. Initialmeetings usually engage in Level III activities only, such assetting a general direction to be specified during moredetailed, Level II work which will be accomplished later.Accountexec 3 is recording the strategy when Director 3,acting decisively (perhaps as the general manager would),initiates a phone call to involve a technician in the meeting.He is not collaborating with the team but abruptly changingthe level of the meeting without allowing any input fromothers present. Technician 6, suddenly joining the meetingover a speaker phone, seems awed by the high status of thegroup. Accountexec 3 would be entitled to be annoyed at suchbehaviour on the part of Director 3. However, Accountexec 3118manages to save the situation by using humour to acknowledgethe difficulty of the task suddenly handed to Technician 6:AE 3 Okay. So part of our strategy is ... (starts towrite)(D 3 has turned to the phone. He connects withTechnician 6 in a distant branch office. AE 3had said earlier he would call T 6 thefollowing morning.)AE 3 (looks up from his notes when he hears that acall on speaker phone is about to begin) Who’sthis? Technician 6? He (ID 3) has been hangingaround with [the General Manager] too long!CD 3 introduces meeting members to T 6.)T 6 I feel honoured -- such a senior audience!*AE 3 (very briefly explains client’s needs.)That’s it. What’s the best solution?(general laughter)D 2 Quick now! Hurry up! (laughs)T 6 (laughs) Do you want the bid too -- I mean thequote on the book?AE 3 (when other laughing voices overlap, speaksmore loudly) We’d also like a detailedconfiguration...D 4 Yeh, can you get that on a fax right now?(more general laughter)Accountexec 3’s jocular manner invites Technician 6 into themeeting, especially when other members join into the hilarity.The humour, as previously noted, also resituates the suddenLevel II request as basically ridiculous, and acknowledged as*Double hyphens ( -- ) are used in all dialog transcriptsto represent breaks and pauses characteristic of orallanguage119such by the majority of team members. Technician 6 isnevertheless being welcomed as a team member for future moreappropriate engagements at Level II, and given a minute tothink, or perhaps to excuse himself, without losing a sense ofhis own expertise.Further in this exchange, Technician 6 does indeed riseadmirably to the situation and settles down to dispensinginformation and giving advice as a sympathetic collaborator.Accountexec 3 does most of the talking for the executives,posing questions and clarifying responses. He shifts flexiblyinto a Level II mode, allowing the technician authority as asource of expert information. The account executive furthervalidates Technician 6’s commentary by using his key phrases.In one interesting example, Technician 6 mentions a “roll yourown” solution. Accountexec 3 echoes back the phrase inmodified form over the next several sentences as a “grow yourown” solution, and finally a “home grown” solution. Thisseems an interesting and effective way to build cohesion withTechnician 6 and establish him as important team member.Accountexec 3 and Technician 6 have collaborated on meaning,and information has been gathered and mutually investigatedfor possible inclusion in the proposal text. However, theinitial Level III objective, that of building a generalstrategy, has been disrupted and the time of several senioremployees, wasted.120One final example of the importance of team building,this time to include the client, shows Accountexec 3 onceagain nurturing a relationship. He reports on Steve, theSundial contact, as a man he genuinely admires and wants toinclude as a voice with vital impact on the proposal contents.Here, then, is audience sensitivity, a collaborative activitywhich appears at all levels to build team relationships:AE 3 The sponsor of this is a manager, [Steve], agood guy, but he’s totally a rebel, ananarchist. He imagines the project as apartnership between [Sundial] and the vendorwith [Sundial] managing it.He returns to his assessment later in the meeting andstrengthens his evaluation of the Sundial contact:AE 3 This guy, [Steve], is a rare breed of cat. He’sa really good guy but he’s a total anarchist.He’s not politically motivated or powermotivated. He’s not going any higher in thehierarchy. He is where he is. He is wherehe’s going to be for the next three years[before he retires]: he just wants to do agood job.Jake, the client’s project manager on the Borden proposal,never attains collaborator status but rather remains aconstant concern to the Cerebellum team as a “loose cannon.”In contrast, Steve, largely due to Accountexec 3’s work, helpsto advise and direct the proposal as a full team member.Finally, the initial Level III meeting maintains a visionof the entire proposal process. In collaboration with eachother, members realize that their general ideas must be121detailed, even verified by Level II, information gatheringactivities:D 3 There’s nothing to stop you from talking to[Jobber] [a vendor currently working forBorden]. Why speculate? We could talk tothem. They’re the guts going into this.They’re interested. See what you can get.Director 3 here urges the team to remember Level II concerns.Although his suggestion seems rather Machiavellian, Director 3does show that he is aware of what needs to be done inupcoming proposal stages. He seems uneasy, though, with hisrole as collaborator. Director 3 begins by advising that theteam (“we”) seek information; he ends by giving theresponsibility to someone else (“you”). Similar behaviour isexhibited by Director 2, also anxious to move the proposal asefficiently as possible into Level I activities:D 2 Give [Manager 1] an example of proposal tofollow.AE 1 (to M 1) For structure, use [Bellum]. Fortechnical content, follow [Alpho] and [Crump].Director 2 is trying to impose an order on the process, andhelp the manager to become one of the team: a group ofcollaborators who do things (like write proposals) in similarways and therefore need models and general specifications. Hesuggests that points of reference be provided for internal andexternal text structure. The new manager will be able tocollaborate with employees not necessarily on the currentproposal by considering their work in previous successfulefforts. Some of the text he encounters will even be recycled122by him, as a kind of “clip-text,” for use in the Bordenproposal.Another instance of overall process awareness comes fromDirector 1. The presentation of the general strategy iscrucial to the success of the proposal, and an ideal approachcan easily be lost in the time pressure of the loomingdeadline. As has been noted above, senior management is givento expressions of ideals, and their hopes for draftingprocedures are no exception. Although Director 1 is awarethat the technical solution will be built along with the textduring Level I collaborations, he is clearly not happy withsuch behaviour. Here he attempts to maintain the boundariesbetween steps in the proposal process:D 1 Let’s decide what we’re going to propose beforewe start to write. That’s reasonable butunusual [for this company]However, such a lock step approach to text production may notbe reasonable or even advisable, since, as we will see, thetext and the solution are built together during Level Icollaborations. Director 1 does, however, maintain a sense of“we,” the group as collaborators, unlike Directors 3 and 2,and thus helps to maintain team relationships.Event B. Level II. Opening MovesOnce a general strategy (Cerebellum as prime or leadpartner with Sauct as secondary partner) is decided upon atthe Level III meeting of the Borden proposal-writing team, the123Directors move off to other company business, leaving theTechnicians and the Manager to begin a Level II informationgathering session. As previously explained, those Level IIactivities in which the intent is to access and gatherinformation tend to occur in small groups of collaborators,most typically two, and are often of relatively shortduration. However, checkpoint meetings where information isconsidered for implications on the proposal effort are usuallylarger. Personnel involved vary in experience from Directorto junior Technician. The hierarchy of authority among themembers of level II teams rests on relative knowledge. Thecollaborator with more information quickly establishes thelead role, but the quality of these collaborations remainslargely pragmatic and neutral in tone, unlike the cut andthrust of the Level III discussions among status seekers.Level II collaborations are quiet and deliberate, remarkablefor activities such as seeking and recording, judging andclarifying. The drama of Level III has given way to practicalmatters, such as careful reading, talking and listening, andof course meticulous record keeping. Level II meetings arealso concerned with the general themes of collaborationleading to text production: team building, audiencesensitivity, schedule maintenance, overall process awareness,and level maintenance.The primary objective of the Level II collaboration is torealize in specifics what Level III made clear only ingeneralities. Immediately after the initial Level III meeting124on the Borden proposal, Accountexec 1 and Manager 1 situatethemselves in Accountexec l’s office, where they begin tocollaborate. Manager 1, as a new hire, uses Accountexec 1 asan information source who knows about the Cerebellum processfor completing a document:M 1 I need to walk through the process.AE 1 We have a deadline of January 28th, Thursday.(begins to write on white board)l. by Tues.[19th] finalize plan and partners --responsibilities 2. tomorrow [15th)content/outline 3. by Mon [18th]- teamresources 4. checkpoint with Director 1 Wed.[20th) 5. writing until early following week --by Fri. [22nd] a copy, by Mon [25th) commentsback, by Tues. [26th) final review. And that’sbasically it.Although information gathering might seem to be a fairlystraightforward activity, obviously Accountexec l’sexplanation is not well organized. This is amelioratedhowever by the written record of his thoughts to aid Managerl’s interpretation. It is interesting to note that whenManager 1 distributes this list as information for others onthe team in an e-mail memo, it is in chronological order withthe comment: “Attached is a tentative proposal schedule that[Accountexec 1) and I developed after the meeting.” Manager 1situates himself as a collaborator with Accountexec 1, andcontinues to establish his own team role as proposal manager.Making meaning is not always as straightforward, however,when the information source is a document, and is open tovarious, more or less advantageous, interpretations. Here we125see a sequence of charting, talking, more charting, reading,listening, more talking, still more drawing and writing beingemployed to access information from the RFP document.However, these collaborators are not merely discovering thecorrect information but rather creating an acceptableinterpretation of the information:(On whiteboard, AE 1 draws his understanding ofthe RFP’s requested hierarchy of relationshipsamong Cerebellum, Borden and Sauct. M 1 demursand draws a different interpretation beside AEl’s. AE 1 re-reads relevant section of RFP.)AE 1 Ah, I didn’t read it that way (modifies his owndiagram: still unlike M l’s). We don’t want(waves at Mts diagram). We’re arm’s length.We want to minimize our risk.Notice also that Accountexec 1 has obviously taken control ofthe meeting due to his authority as information source. Heknows about the client audience. And although he is preparedto listen to and observe Manager l’s ideas, his interpretationof the RFP is clearly not to be further disputed.These conversations indicate as well a general theme ofcollaboration: keeping a comprehensive view of the proposal.Manager 1 clearly bears much of the responsibility forcompleting the proposal on time, and will do a significantamount of drafting. His request for information on a timelinefor Level I work is therefore not surprising. Then too,Manager 1 must also be concerned about the business strategyand keep in mind the general directions prescribed by theLevel III meeting. He moves back and forth rapidly between126these concerns which Accountexec 1 tries to treat largely asrequests for information from a new employee, resisting areturn to the open egalitarian banter of a Level III debate:M 1 [Director 1] said we should do content beforewriting.AE 1 Do a table of contents, then we’ll fill in theoutline.M 1 Okay.AE 1 I’m not going to be here tomorrow.14 1 This is a high-risk project --AE 1 The owner was told that this was problematic.But we will “minimize your risk”. They knowthat.14 1 So in the end, they take the risk....AE 1 Our managing reduces risk.14 1 What happens two years down the road when theyneed to spend another million?AE 1 I don’t know.Further elements of a level II meeting are present here.Manager 1 attempts to reinforce his view of the proposalprocess, reminding his collaborator about an idea flowing froma Level III dictate earlier voiced by Director 1. Accountexec1 suggests outlining as an appropriate step in the textproduction process, and perhaps as a way to respond toDirector l’s concerns. However, he also tries to shut downthe conversation in several ways: giving quick answers,indicating that he will be busy tomorrow (perhaps also hidingbehind his status as a more senior employee with otherduties), and finally refusing to answer a difficult question.127Manager 1 however, persists, indicating his awareness of andconcern for items which he does not feel were satisfactorilyresolved at the Level III meeting. Accountexec 1 clearlywants to resolve his difficulty by engaging in a rehearsal ofthe rhetoric needed in Level I to address the audience’sconcern for risk.Another common theme evident in this Level II meeting isthe maintaining of Level II activities. As noted above withManager 1 and Accountexec 1, Level II meetings can be strainedif conversants do not share the same view of the level andthus the objective of the collaboration. Accountexec 1encounters such a reaction once again when looking forinformation in another Cerebellum branch office. He connectswith Technician 3 on the phone:AE 1 We’re not going to bid Azure. The consensuswas to go [Sauct] . It looks like [Jobber]doesn’t have the capability. [Cerebellum] willtake responsibility for the project.T 3 Very risky for us. I’m reluctant. [Sauct] istechnical plus. What do we have to gain....AE 1 It’s been decided. But thanks.T 3 [Technician 11 has the files [you’ll need]Technician 3, like Manager 1, wants to debate in LevelIII fashion. But whereas Accountexec 1 seems to feel obligedto keep talking to Manager 1 in a role of informationdispenser, from Technician 3 he needs information (Level II)and is not prepared to wait any longer or debate with him(Level III). He is engaged in level maintenance——indicating128to a collaborator what behaviour is helpful at this level. Hetells Technician 3 quite abruptly that the “decision has beenmade”—a Level III job since completed, reminding Technician 3perhaps that his comments are inappropriate in a Level IIinformation gathering session. Accountexec 1 does, however,try to soften his comments by thanking the technician, tomaintain a positive relationship with at least superficialpoliteness. Technician 3, just as abruptly, indicates thatthe information sought is already in the branch office withTechnician 1. He gives in to Accountexec 1, conscious perhapsthat he is not in a position to argue further, although he hasknowledge which should be a source of authority at Level II.As we have seen before in Level III collaborations, thecorporate status hierarchy has been reinforced here byAccountexec 1. The collaborative activities possible areimpeded by such level jumping. A potentially useful Level IImeeting has been derailed by a jump to Level III.Days Two and ThreeDays two and three are consumed with various meetingswith the Borden client, the client’s project manager, andvarious possible vendors who are possible partners forCerebellum’s bid. Most of these meetings took place out ofthe office and were not made available for data gathering, andthus are not reported here. However, the net result for theproposal was a large amount of information.129Day FourBy Day Four, the proposal team has amassed a significantnumber of facts and opinions about their competitors on theBorden proposal, and more about the needs of their Bordenclient. By the end of Day Four, Manager 1 has made notes tooutline a new business strategy, and Technicians 1 and 2 havedrawn up a list of hardware needed to complete the project.Event C. Level III. Changing HorsesAlthough the initial Level III meeting (Day One, EventA), established a general direction, Level II informationaccessing, gathering, and consideration results in amodification of the original plan. The following conversationtakes place in a hallway, where a group of four senioremployees (Director 2, Accountexec 1, Accountexec 2, andManager 1) hold a hurried Level III meeting. These executivesare trying to find a new strategy which will counter the riskof the competition’s aggressive pricing. Emotionalvolatility, a quality of Level III collaborations, is apparentin the conversation. Clearly, the battle has begun.Decisions, even Level III ones, must be accomplished withhaste. And so the collaborators make a daring leap,accompanied by the warnings and reassurances characteristic ofLevel III collaborative activities:AE 1 We could make all three [competitors] part ofthe proposal.AE 2 That would ensure [our involvement]130D 2 They [would] all want us to be prime and takerisk...AE 2 ... so put us in charge...D 2 but vendors won’t say that [we will take allrisk] and [Azure] will put in a separate bid...AE 1 We’ll discount...D 2 [the General Manager] wants the business.Cerebellum has a new course of action—but is it a risk or anopportunity? Or perhaps both? The decision is clinched bymention of the general manager’s desires—he is clearly a teammember, a fellow collaborator, if an absent one. Apparentlyalso, the general manager is first in status amongst thesecollaborators, and his is the authority in a Level IIIcollaboration.The general themes of collaboration are present in themeeting as well. Maintaining team relationships becomes acrucial factor. The tension present in other Level IIImeetings is obvious. Although Director 2 slips into the roleof the (reputedly parsimonious) client/owner, Manager 1, whonow seems more sure of himself, is unimpressed and reacts asan equal. He feels free to reject Director 2’s view that therelationships of everyone, client and partners, must beconsidered at the present time. As we will see, theyencounter some difficulty in deciding on the primary audiencefor the proposal. Director 2 and Manager 1 do not seem tohave in place an understanding of a very basic rhetoricalconcern—whom are we communicating with, and what is theirprimary interest? Director 2 hastens to maintain a positive131relationship with Manager l—--”your points are valid”—reminding him perhaps that a Level III meeting must take abroad view of the possible consequences of their strategy:(0 2 roleplays client’s probable rejection of ahigher price to M 1, who does not respond orreact in any way.)M 1 [Borden’s project manager] doesn’t know who todeal with.D 2 This situation isn’t fair to our possiblepartners. We have to make a decision.M 1 Look at it from the client’s point of view.D 2 No -- I mean your points are valid, but I’mtrying to look at the psyche of people [otherthan the client] involved.Worthy of note also is that Director 2 has earlier remindedthe group of upcoming Level I issues: “We’ll have to do a lotof word-smithing on this one.” His view extends not only torelationships, it seems, but also to a comprehensive view ofthe entire collaboration process as a venture which leads to adocument.Event D. Level 11. Who’s on first?As the pace quickens, it becomes apparent that every newpiece of information is causing ripples of concern amongstteam members. Two examples will show how Level II meetingsnot only gather information, but also consider the effect ofthe information on the decisions made at Level III. Thetechnical architect on the proposal effort, Technician 1, hasspent several days acquiring and evaluating information. His132remarks to Manager 1 show the ongoing competition betweentechnical and sales concerns which permeates all levels andoften threatens the ability of the collaborators to workeffectively together:T I. I can’t see how this [the Level III strategydecision] will work. How will we maintain ourcredibility with vendors? It feels like asales decision.M 1 Oh really?T 1 Whose idea was this? [Director 2] ‘s?M 1 I don’t know.As a technician, he finds himself uneasy with a Level IIIdecision based, he suspects, on a sales motive without dueregard for the difficulty of maintaining relationships withtechnical personnel in other companies. His role as aninformation gatherer, though, should be to ask questions whichwill enable Level III decisions to be enacted. This event isan example of level jumping. In the event described above,Technician I seems to feel vulnerable in his Level II role.He strikes out at a representative of the Level IIIcollaboration which has resulted in a decision which he mustimplement.In much the same manner as Accountexec 1 on Day One,Event B, now Manager 1 finds himself uneasy with the Level IIIaggressive feeling of this brief encounter. Clearly, hecannot make a Level III decision. However, Technician l’scomments are not Level II information gathering questions(ironically, in similar fashion, Manager l’s comments on Day133One, Event B were also uncomfortably forceful in nature duringa Level II meeting). Although Technician 1 is concerned withhis own relationship with his information sources, he does notseem to be concerned with his relationship with Manager 1during this event. Such behaviour may result from unresolvedLevel III tensions, and may lead to less commitment amongstand between collaborators responsible for drafting proposals.Event E. Level II. Decisive ActionThe second example of a Level II encounter contrasts withthe first. Here, Manager 1 is sharing information withAccountexec 2. There are multiple effects of one piece ofpowerful information. First, relationships are beingmaintained. The owner will be informed in a Level II meeting,which seems intended to keep him feeling like a member of theteam. Such a meeting will be respectfully initiated by asenior executive. The project manager at Borden must also beconsidered. Finally, Accountexec 2 and Manager 1 are seenhere to be collaborating smoothly. Accountexec 2 does notinform until asked. Manager 1 has correctly evaluated theinformation about the software and has informed someone whowill be able to direct him further:M 1 How can we guarantee software if it’s not up[available] yet?AE 2 I wasn’t aware that was happening. Thatchanges things. We need a meeting with theowner to clarifyM 1 I need some advice.134AE 2 Get Director 2 to call [the owner] . He caninclude [Jake] or not. We don’t want to beseen going behind his back. We want to providea solution, not just get businessIt is interesting to note also that the sales person,Accountexec 1, does react to an important technical issuedespite Technician l’s fears that the sales agenda takes undueimportance in the proposal. The meeting between Accountexec 2and Manager 1 shows information sharing, gathering andevaluating: rapid, pragmatic and unemotional. It also revealsa hierarchical collaboration--the more senior person directingsubsequent activities.Day FiveDay Five begins with collaborators still gathering,evaluating and considering the implications of informationboth to the building of a technical solution and a proposalwhich presents that solution. By the end of the day, Manager1 has completed a Table of Contents for the proposal. Thelist of topics will serve also as the headings for a proposalworkbook. Manager 1 has also reviewed the proposal plan withthe support staff so they will be ready to schedule and acceptdata for word-processing work.Event F. Level II. Under ConsiderationThe following meeting involves Manager 1, Technicians 1and 2, Director 2, and Accountexec 1, and has been called byManager 1 as a Level II information checkpoint meeting.135However, he initially has difficulty getting Director 2 andAccountexec 1 to co-operate with his agenda and collaborate atan appropriate level. In fact, the opening moments of themeeting bear many of the characteristics of a Level IIImeeting: banter, conflict, predictions and glowing visions.Accountexec 1, perhaps anxious to ally himself with Director2, supports his attempts at humour. Manager 1 at firstresponds to the bantering, but quickly tries to re-establishthe meeting as an Level II information session. As anewcomer, he is under considerable pressure to yield to thegreater status of Director 2; however, he still assertshimself as the chairperson in a Level II meeting whereauthority usually rests with the information givers. He maybe overly anxious, missing the opportunity to engage in teamrelationship building:AE 1 You know what [the competitors will] do though,they’ll guarantee a response time but theywon’t guarantee the platform...M 1 OK, can we...AE 1 . . .that it’s going to go on...M 1 Ok, can we ... can we ... let’s not resolvethis today right now. We’ve got 6 minutes.[before D 2 must leave]D 2 Three.M 1 Three minutes.D 2 Two.AE 1 Two! (laughs)136M 1 The other thing we’ve got on our plate to getout is this generic list of equipment bylocation. How’s that going?Level II meetings are generally very pragmatic affairs, withrelations being maintained in rather more indirect ways thanin the drama of Level III. Note also that the technicians donot participate in the struggle to direct the meeting, perhapsdue to an awareness of their relative lack of power, orperhaps due to their unease with an unfamiliar Level IIIsituation.Director 2 reminds Technicians 1 and 2 of the timepressures the proposal faces, and the interdependency of theproposal writing process: examples of schedule maintenanceand team maintenance. Nothing can be finalized until thecosts are known, and the costs cannot be known until thesolution is built. The tension between due care andattention, and need for speed is apparent. Technician 1 mayresent the need for cost information when he is stillsearching for the right technical solution:D 2 OK. So we’re going to have a network pricedout, rough cut, either today or tomorrow --T 1 Tomorrow.D 2 ‘kay, smart man. (laughs)Director 1 seems to acknowledge Technician l’s refusal to berushed while trying to gather and consider information, andshrugs it off with a laugh. He is trying unsuccessfully tomaintain the proposal building schedule which conflicts withTechnician l’s role in the maintenance of technical quality.137A Level II checkpoint meeting also involves reports ofother information gathering activities. Technicians 1 and 2indicate a collaborative working relationship both within andbeyond this particular meeting in gaining and sharing relevantinformation. Joint reports by collaborators are important atcheckpoint meetings. Here we see an example of proposalcontent being reported without much success by Technician 2.Although these collaborators were later observed working infriendly fashion together at both Levels II and I, they do notco-operate in this event, perhaps due to differing notions oftheir team responsibilities in the meeting and to the LevelIII representatives present. Although Technician 1 is seniorto Technician 2, he apologizes for his interruption, at leastinitially. However, he then interrupts again, perhaps becausehe has decided that the others at the meeting want estimatesof completion and not a detailed technical listing ofeverything they have accomplished:T 2 That’s what we’ve been working on. [Technician1] is going to talk to [Telit] to find outexactly which...T 1 That’s ... that’s ... sorry.T 2 Where the concentrators are going to go and...T 1 That’s very close [to being finished] . I meanthe table itself will be populated.During the rest of the meeting Technician 2 continues tocontribute freely, apparently undeterred by this interruption.She refers to Technician l’s plans, and otherwise attributesthe work to both of them—”we’ve been working”. Her use of138the plural pronoun is echoed by Technician 1 throughout themeeting.Perhaps because the Borden proposal is now only eightdays from deadline, much of this Level II meeting also showsparticipants maintaining an awareness of the proposal writingprocess as a whole, thereby helping to maintain the schedule,especially with regards to Level I drafting activities:D 2 So the only thing left from a pricingperspective is the services and somebody shouldbe working on that, and, we should have --M 1 That’s the next questionD 2 -- we should have some of the proposal finishedthis week -- some of it.(AE 1 laughs)D 2 Preferably more than less ... (smiles)Director 2 is definitely trying once again to hurry the teamalong and maintain the schedule. Ironically, he andAccountexec 1 pause to banter in Level III fashion once more,interrupting Manager l’s attempts to discuss the informationstill missing and thus derailing maintenance efforts. Later,Manager 1 returns to the drafting discussion, indicating thathe too is aware of its importance. While Manager 1 pointsforward to Level I, now it seems that Director 2 has neitherthe patience nor the time to discuss Level I activitiesfurther. As befits a Level II meeting, however, he does givea piece of information, an example of level maintenance. Thesuggestion, however, indicates his belief that Manager 1should be doing the drafting. Director 2 is avoiding a fully139collaborative role at Level I (a stance which will later proveto hamper schedule maintenance)14 1 The other action item here is the writing ofthe proposal, right?D 2 A pretty important part.14 1 So we need to start dealing with that.D 2 Yeah, yeah.14 1 I can start outlining content with [Director 1)at 10 o’clock.D 2 Get him to cite good quality examples and startcutting and pasting them. (he looks at hiswatch)Although he obviously believes that the proposal drafting isimportant, Director 2, it seems, does not want any involvementin actual text drafting. He even suggests that Manager 1would be well advised to cut and paste, that is use “clip-text,” rather than compose original text, at least at thisstage.Technician 2 also indicates that she is aware of draftingduties, and indeed already involved in Level I activities.She even recites, or perhaps rehearses, sentences and ideasshe and Technician 1 are using in their draft of the technicalsolution during what seems to be a Level II meeting:T 2 .... so we’re saying “put in a four port pad,because you have a choice ... so we’rerecommending four then you can have an optionto upgrade it . . . .“Here Technician 2 engages in content specifying and textrehearsing. Text rehearsal would perhaps be better suited to140Level I where she would have a collaborator focused onappropriate and precise working. She does seem to wantDirector 2 to understand what is being recommended veryspecifically to the client in the proposal, which will perhapsmake her drafting more confident. She appears to be seekingreassurance, which she otherwise usually obtains fromTechnician 1 (seen in various glances to check his reaction toher statements). Technician 2 may feel unsure of her role onthis team, or of her authority as a technical informant andpresenter of the technical solution.Such Level I conversations are especially interestingduring this Level II meeting since Manager 1 indicates he isstill concerned about Level III decisions:M 1 .... we’ll start getting requirements andparameters and start to do some of that scopingof the project that hasn’t ... you know thatneeds to be done soThis is another example of level jumping arising from asense of vulnerability. In general, this Level II meetingshows that information gathering is correctly placedbetween strategizing and drafting. Facts, it seems,mediate between high-level activities such as scoping(that is, defining) a project and rendering it into text.Information changes everything from the approach to theproposal to the very words which will describe theapproach.141Event G. Level III. Quick Change ArtistryAlthough I was not able to record this meeting, lateon Day Five the most senior executives met to considermodifying the strategy in the Borden proposal once again.The General Manager directed the meeting in a small officebarely containing the participants all eager to argue andscribble diagrams on the whiteboard. Eventually, onepossible partner emerged as the definite front runner.Azure would be presented as the partner/vendor of choice,with two others offered as alternates. The strategy, afull five days into the writing process, had finally beenresolved. Clearly these collaborators needed muchinformation in order to arrive at a final formulation.The sometimes chaotic nature of Level III meetings seemsto make reaching an agreement somewhat problematic, giventhe tight time lines of proposal preparation.The Level III meeting late on Day Five also marks the endof the early phase of the writing project. A final decisionon strategy means that energy can now be devoted to otheractivities. The team has drawn together in common cause atlast--at least on this issue. More shifts in direction,however, would come later.142CHAPTER SIX: PRESENTATION AND DISCUSSION OF MIDDLE EVENTS“Curiouser and curiouser!” cried Alice(Lewis Carroll, Alice’s Adventures in Wonderland)A general strategy was finally settled at the Day Five,Level III meeting of the Borden proposal-writing team.Cerebellum would test three optional partners for the bestsolution to Borden’s hardware needs. Azure would be testedfirst, and if proven adequate, adopted without further testingof other vendor hardware. Much information has been gatheredwhich prompts that decision by Cerebellum’s senior staff.This information must now be further considered by allinvolved for drafting into text. Thus begins a series ofLevel I drafting sessions (though Level II informationgathering also continues). Unlike Level II where informationis gathered and then checked, drafting begins with a checkingphase of reading, rehearsing, reviewing and justifying text.Checking is followed by building text in detail. Aspreviously explained, Level I collaborations tend to occur insmall groups of collaborators, most typically two (or evendrafters working alone); however, meetings where draftingissues are checked are usually larger. Personnel involved intext drafting are managers and technicians.The primary objective of a Level I collaboration isdrafting text. These collaborations are intense and focused,remarkable for the building, checking and merging of varioustextual possibilities. Activities are often warranted by143referring back to the audience and purpose articulated atLevel III meetings, and to the facts gleaned during Level IIwork. Level I meetings are also concerned with the generalthemes of collaboration leading to text production:maintaining the schedule, working within appropriate boundarylevels, and overseeing the entire proposal writing process;and maintaining the proposal writing group, building andsustaining member roles and inter-relationships, and includingthe views of the perspective audience.Level I collaborations gradually result in more and morefine-grained and sustained text—points become sentences andsections as collaborators continue to meet. Finally, muchtext will be composed in physical (but not actual) isolation:even writers who seem to be alone are working with ideas fromothers on the team.Some of the proposal will be lifted and adapted frompreviously written documents rather than originally composedfor the present task. Throughout the proposal, writers turnto key extracts of text to help them maintain the proposalschedule, focus on the audience, and even write in a companyapproved style. Text borrowed from other documents is morequickly assernbled.than is text which must be composed word-byword. Also, such “cut and past&’ text, it seems, is assumedto be adequately written and presented given its previouslypublished status. Recycled ideas have already been acceptedby a client audience, either by this client in particular, orby some other satisfied party (another client in the case of a144technical proposal, or the corporate executives in the case ofa Cerebellum company profile). Finally, “clip-text” serves tokeep writers employing company approved prose style as theywill be writing words around previously approved text or usingpreviously approved ideas.Day SixBy the end of Day Six, collaborators are working to fillthe workbook binder. The binder is filled with a series ofsheets, titled with hand written headings. These form themajor sections of the proposal and a list of appendices.Technicians 1 and 2 are using clip-text and writing originaltext to complete the section called “current environment,”that is, what Borden has in place at the current time. A listof writing jobs is also complete.Event H. Level I. Problems, ProblemsHere we see Accountexec 1, Manager 1, and Technicians 1and 2 meet to begin drafting the proposal. Their goal is tocheck out ideas and their presentation of those ideas byengaging in various collaborative tasks unique to Level I:composing aloud, thus checking by rehearsing in front ofothers, and justifying choices by referring back to levels IIand III. Level jumping and task avoidance seem to interferewith progress in this Level I meeting.This event begins with Manager 1 struggling to establishthe meeting as Level I and maintain the proposal schedule(both ongoing collaborative themes), but encountering145resistance from Accountexec 1. Accountexec 1 engages in taskavoidance here perhaps as a way to attempt to escape fromLevel I drafting. He tries to establish a role outside of theone which the team needs to maintain their schedule--that oftext composer. Accountexec 1 seems to feel that he should beexempt from this activity (except perhaps, as will be shownlater, as an reviewer) . The reaction of the others,especially the laughter from the technicians, indicates thatthey feel his behaviour is inappropriate (though perhaps notunexpected). A power play suitable to a Level III gathering,where disagreement over roles is expected and can be debated,is not helpful to the collaborative work of text production atLevel I. Manager 1 reminds him of the ongoing concern ofschedule maintenance--a time line which will not proceed asplanned without co-operative action:M 1 I want to talk about content and maybe talkabout who can help me fill in the blanks orwrite certain parts of this for me.(AE 1 points to T 1 and 2)14 1 (to AE 1) Well, I think you can help too.T 2 (to AE 1) Don’t look at me.14 1 I’d like to see -- if that’s all right with you-- I’d like to see you participate in some ofthis.AE 1 Yeah, some, but14 1 This has been yours from the start.AE 1 No, I know. I can do some but -- I’ve got a lotof things happening. (laughter T 1 and 2) No,seriously.T 1 You’ve got a course today, right?146AE 1 I’ve got a course today and I’ve got anotherdeal that I have to finish by next week. Newbusiness. But I’ll do whatever --M 1 (interrupts) Let’s talk about our deadlines.Of interest here too is Technician l’s acknowledgement ofAccountexec l’s other responsibilities, a way to maintain teamrelationships by allowing Accountexec 1 to maintain hisdignity and sense of belonging in the group. Team membersbear responsibilities at Level I to build text (as well asdebate it), and Accountexec 1 has been forcefully but notunkindly reminded of his role.A second example of a conversation which disrupts thisevent as a Level I meeting is once again an attempted shift toLevel III. In this debate we see Technician 1 engaging indebate over the strategy, rather than maintaining the Level Iprocess. Clearly, Technician 1 is still unhappy with thestrategy especially as it seems to him that it may involvebehaviours which will jeopardize his relationships withvendors (see also Event D) . He is jumping to Level III,trying to establish a different vision (and challenge the onethey are encapsulating in the text) of the team which includespossible vendors as fully informed members:T 1 It’s a weighted approach --AE 1 Yes.T 1 It’s weighted to [Azure]M 1 Only because of other factors.T 1 Right. But it still is weighted. I’m justtrying to clear this up. The initial147understanding from the vendors’ point of viewwas --M 1 We discussed this and [Director 3] said: “Don’tstir anything up.”However, this is not an issue which can be resolved at a LevelI meeting (although the text under production cannot help butremind him repeatedly of their strategy) and Manager 1 remindshim of the high-level decision, already made, whichestablished the general approach. The debate continues.Eventually, however, Technician 1 begins to accede to theplan:AE 1 Why have [the client] spend more moneyunnecessarily?T 1 Right. So it puts a lot of the emphasis onAzure to get it right.AE 1 That’s right. Yeah. (long pause) Clearly it’sthe one that best suits the client and that’sthe driving factor here.T 1 Maybe not technologically but --AE.1 -- from a business perspective.The ongoing competition between technical and sales issues isalso evident here. Accountexec 1 reiterates that a businessstrategy is best for the client, one which takes technical andsales considerations into account. Technician 1 is remindedthat he must take a broad view. This strategic issue,however, might have been resolved in a Level III meeting.Maintaining an awareness of the whole process, ratherthan level jumping, is a legitimate collaborative activity andfunctions to drive the proposal forward. While discussingvarious topics in the proposal and deciding what will be found148under each heading, it sometimes becomes obvious tocollaborators that their Level II information gathering isincomplete. In such cases, speed is imperative in obtainingmissing data and an experienced team member is an invaluableasset. The authority of Level II, based on knowledge, isevident here, both between Accountexec 1 and Technician 1 andthen between Technician 1 and Technician 2. The collaboratorwho knows more takes charge. This brief interlude quickly andsmoothly handles the information deficit. Technician 1 seemsto know the answer to his complaint and is perhaps remindingthe team that his role is not without difficulty:M 1 Can you have that info by the end of the weekor Monday?T 1 Well, we can try. I mean, you know we have nocontrol about that part of things.AE 1 Well, what you can say to them is --T 1 “If you’re serious about the business --“AE 1 “we need it by tomorrow. If it’s not in bytomorrow --“T 1 “then we won’t use you.”AE 1 “We won’t use you.” Right?T 1 Right.M 1 You’re going to handle that?T 1 We’ll split it, Technician 2 and I. (to T 2)I’ll call half, you’ll call half.T 2 (writing) Yeah.(discussion returns to headings and subheadingsneeded)149Technician 1 maintains his position as technical expert hereby assigning work to Technician 2, but he also shares the loadand includes her as a valuable team member by doing the sametasks. Accountexec 1 engages in schedule maintenance, in afascinating example of collaborative sentence building, byreminding Technician 1 of the solution to his problem. Hedoes not, however, offer to take on any of the informationgathering work and maintains his pose as a somewhat distantteam member.Level I activities include the collaborative resolving ofissues such as what should be included under each heading, andin how much detail, and who will actually compose the text.Discussions at checkpoint meetings must centre on generalprinciples but will also deal with specific ideas. One suchgeneral issue is the assignment of sections which individualteam members will draft:M 1 Obviously the executive summary will be writtenlast.AE 1 I’ll do it!!!T 2 (laughs) Oh yeah, he’ll just cut and paste.T 1 He’s already got a canned version.AE 1 I’ve got one somewhere!M 1 Are you serious?!AE 1 No. I’m just kidding. Seriously, I will writeit. ‘cause I have a specific approach.M 1 (vaguely, looking down at notes) Have you? Wereviewed this outline with the client.150Although the humour here has an edge—the technicians seem tofeel that Accountexec 1 must be unfairly avoiding yet anothercomposing job——the manager remains focused, avoids beingsignificantly drawn into the exchange, and once more maintainsthe collaborative level. The account executive tries to takethe opportunity to re-establish himself as a collaborativeteam member by taking on a composition job, thus maintainingteam relationships, and perhaps indicating by his laughterthat he is aware that the team is not happy with his generallack of participation.The team reinforces their expectations of Accountexec 1as a Level I collaborator later in the same meeting, and heshows more sensitivity to his role as drafter. Accountexec 1shows considerably more sensitivity to the collaborativeeffort. He graciously accepts more drafting work and stillmanages to maintain his role as sales leader. This is, afterall, his account and his ultimate responsibility, and he iskeeping a broad view of the proposal writing process.Accountexec 1 also rehearses his composition, perhaps for teamapproval. This is a behaviour typical of Level I and in thiscase the rehearsal does indeed engender a response:M 1 We want to reiterate in this section ourunderstanding of his needs....T 1 I think that’s an [Accountexec 11 category.What do you think?CT 2 laughs)AE 1 I usually refer to it as a business problem.(smiles) Thank you. (writes in his notes)151T 1 My pleasure.M 1 You’re going to write the whole thing?AE 1 Yeah. Well, not get into thehardware/software... [but] how they’re “lookingfor a hardware platform blah blah blah forefficiency of operation”M 1 Are you just going to re-state what’s in theRFP? Or are you going to commentAE 1 We don’t get into too much detail at the minutelevel but rathr keep it at a high level so wecan always get back to the business issuesManager 1 pushes Accountexec 1 gently along in his efforts attext production, maintaining the team relationships whilereminding him that he should be expanding on the RFP ratherthan merely echoing it back. One of the challenges of clip-text would seem to be how to enter it into a new text.Interestingly, Accountexec 1 has not said that he would simplyrepeat the RFP’s phrases, but Manager 1 is gradually learningthe strengths and weaknesses of fellow team members andimproving his collaborative skills. He can also, perhaps,hear the echo of the RFP and is sensitive to it as a text ofconsiderable presence in the formulation of the proposal.Justification of text is another Level I activity. Oncecollaborators begin to rehearse wording, then they are oftenchallenged (as above) or asked to justify their approach. Aswe have seen, Accountexec 1 often claims merit for hiscomposition based on his notion that much of the document mustbe very general—broad principles rather than technical andprocedural details. Despite continued reiterations, however,152the other team members resist his direction. The team returnsto this struggle again and again, wasting time and failing torecognize that their continuing debate threatens schedulemaintenance:14 1 (to AE 1) So you’re going to write this. Theirneeds.AE 1 The business problem ... then after that we say“here’s how to do it”. But keep this at a highlevel. (T 1 smiles and laughs looking at AE 1)“[Borden] has selected blah blah blah and nowthey’re in the process of selecting....”That’ll do it.14 1 I understand you are trying to steer away fromspecifics .... but --AE 1 why highlight if they haven’t asked for it?T 1 He asked for very specific things.AE 1 (reads from RFP, then --) That’s it. That’s allhe’s asking for.The RFP makes a powerful source of justification forAccountexec 1, and he does seem to win this round. Theargument, however, appears to be based in an ongoing debatebetween detail (as desired by the technicians) and generality(as desired by the account executive). Sales needs andbusiness case, it seems, have become inextricably entwined inthe minds of the technicians. Their problems with Accountexec1 as a collaborator, perhaps, have begun to affect their senseof the business case which must be developed in the proposal:a business case consisting of sales and technical attributes.In similar fashion, the Sundial group also discussesgeneral and specific ideas for inclusion in the proposal.153Controversy arises. Many of the features of the heatedexchange which follows bear remarkable similarities to theBorden situation. Once again, we have two technicians whowork closely together and reinforce each other’s ideas. Inthis case, the attack is focused directly on presentationalability, rather than somewhat more obliquely on theinterpretation of the RFP, as in the Borden case:AE 3 You’re a goddamn lousy writer.(general laughter)T 7 No, it’s already written.T 5 Did you read the goddamn notes or not?AE 3 But it’s all in detail. I’m looking for majorthemes. I don’t want detail. What are thethemes? You keep burying me in detail. So theycan get to the image enabling much quicker thanwith anyone else. Okay. So where does it saythat?T 7 So then we’ve got a definition of themedifferent. We put that all under theexplanation of productivity.AE 3 Swing your thinking around from thinking aboutthe solution to thinking about what’s his name,what he’s going to buy this for. What’s goingto make him buy this system, from us?M 2 For image enabling.AE 3 That’s all just weasel words to him.In Sundial, as in Borden, the manager tries to support thetechnicians, once again to no avail, in the face of accountexecutive resistance. And, most forcefully, Accountexec 2bases his argument on an analysis of the audience, a point ofview which he implies the technicians have ignored.154Accountexec 1 does not offer any rationale for hispresentation preferences, although audience sensitivity islikely the primary reason as well.Much of the collaborative text building and checkingduring Level I drafting surrounds the selection of informationto use and the rehearsal and justification of proposal text.Ongoing tasks which maintain the process include maintainingthe level, even after an attempted jump to another level. Theauthority in Level I collaborations is based on skill inpresentation, not knowledge of facts. To maintain the level,Manager 1 must divert the behaviour of Technicians 1 and 2 whowant to impart knowledge about the computer system in placerather than focus on the presentation problem:M 1 Are you going to price options?T 1 That becomes very time consuming.T 2 Yeah!T 1 You have to price wiring .... (continues tolist items)(T 2 begins to explain in detail the client’scurrent technical configuration)M 1 (interrupting) [I] Understand the problem. Justlooking at this from the guy getting theproposal.T 1 Well...M 1 And he says “Okay, this is my base price. NowI’ve got all these options. Should I take it orshouldn’t I?”T 1 Well, we can ...M 1 How can we help him with that decision?T 1 We can give him a ballpark figure.155AE 1 Not detailed ones ... so we can say “Based onyour standard, this is what you can expect topay.”14 1 (to T 1) I like your idea of providing aconstant.T 1 Yeah, we did that before [on another occasion]Together, the group works out a way of selecting and offeringinformation in the text. Of interest too are their attemptsat maintaining team relationships. Although Technician 2 isnot directly acknowledged for her value as a skilledpresenter, her technical problems are acknowledged.Technician 2 has a particularly challenging job as acollaborator: female, much younger, and much less experienced(see Day Nine, Event R. A Woman’s Place). The technical pointof view combined with valuable presentation ideas, aseventually expressed by Technician 1, is validated by themanager. The technical contribution is acknowledged as usefulto the drafting decisions, here rehearsed by Accountexec 1.The text is justified by him because it is appropriate for theintended audience (Borden) and its expec1ations: audiencesensitivity is an ongoing collaborative task. Technician 1also demonstrates audience sensitivity by validating the textstrategy from previous experience with the client.The team continues to work fairly smoothly during thismeeting. They maintain the process by accepting their textbuilding roles and thus supporting the team. Although thegroup is still doing some rehearsing of text, the exchangebelow focuses mainly on role building and team maintenance.156As collaborators, each member accepts responsibility fordrafting assignments which they can best handle. Of noteespecially is the close working relationship betweenTechnicians 1 and 2, who almost always speak of themselves as“we.” Perhaps aided by their collaborative experience, theyalso offer aid freely to Manager 1 who will do first drafts ofmany sections largely on his own:14 1 We’ll say “Here’s what you need to do . . .“Whatever. Can you (T 1) help me in that? I’lltake the lead.T 1 Okay. Okay.(M 1 reads the next heading from proposal outline)T 2 That goes with technical strategy -- so we [Tiand T21 may as well do that.****(M i reads next heading)T 1 It’s a group thing.14 1 I can draft something.T 1 And we’ll fill in.14 1 (records roles) Okay. I’ll draft.AE 1 Just put beside it “H-chart”.14 1 “H-chart”? What’s the “H” stand for?AE 1 That’s the way it looks.14 1 Oh. (pause) Oh! Okay. (gestures in an H shape,smiles)Accountexec 1 does not offer to draft text, maintaining hisrole as a senior employee but avoiding a task. Such behaviouris notable given his ultimate responsibility for the157proposal’s success. He limits his contribution here to apresentation suggestion.Manager l’s lack of experience with the company isobvious here in his problems with the “H chart”. Visualpresentations of strategy abound in the proposal writingeffort at Cerebellum. Accountexec 1 does not offer to buildthis chart, however, despite Manager’s lack of familiaritywith it. Manager 1 in fact does not ask what might be in sucha chart, perhaps due to embarrassment or perhaps due to thehighly focused quality of Level I text-checking sessions whichseem to be characterized by a relatively rapid, straight aheadapproach. At this point, collaborators are considering alarge number of headings which represent topics for possibleinclusion in the proposal. Each topic is discussed to ensurea common understanding, and then assigned as part of the textproduction work. To maintain the schedule, Manager 1 mustmake sure the meeting stays on track. He seems hampered byhis own lack of in-depth understanding and the reluctance ofAccountexec 1 to assist in establishing time parameters.Level maintenance does not seem to be a problem duringthis part of the meeting, which does, however, feature someevidence of overall process awareness, and role maintenancewhich will help to sustain the process. This collaborativeteam has members skilled in different areas. AlthoughTechnician 1 here seems willing to accept responsibility for“pricing” (that is, the amount paid by client: cost plusprofit), he is relieved of that duty by Manager 1 and158Accountexec 1. Cost-gathering is a part of Level IIactivities largely accomplished during the proposal effort bythe technical staff, however, Level III meetings of seniormanagement dictate the price which will be quoted to thecustomer. This sorting out of roles however, seems todistract Accountexec 1 and he jumps into Level II, drawingTechnician 1 along with him. Manager 1 steps in to maintainthe Level I activities by moving the discussion back to theoutline: “Let’s get --“ back to our current work, perhaps:M 1 (reading from outline) Pricing.T 1 What? Ballpark? Estimate? Fixed price?(general discussion, then --)M 1 You guys do the data gathering.AE 1 You do the costing ... we’ll do the pricing.T 1 Is [Director 1] going to review this document?M 1 Yeah. And we’ll want [Director 21’s input.AE 1 When you talked to vendors -- did they give youeverything?T 1 No. They’re not sure it’s right for them [togive prices]AE 1 If they don’t, call [an alternate source]T 1 We’ll fax them.M 1 Let’s get -- let’s talk about timing.There seems to be always a delicate tension between keepingthe entire process in mind, thus maintaining an awareness ofthe different levels, and avoiding the urge to jump betweenlevels, slowing the process and not accomplishing tasksimportant at the current level of engagement.159Presentation continues to be the main emphasis of theLevel I collaboration. Arrangement and diction are tworhetorical concerns repeatedly addressed. This conversationdemonstrates clearly the skill based authority incollaboration which characterizes Level I. Writers worktogether offering and accepting suggestions for the mosteffective presentation of ideas regardless of seniority. Herewe see Accountexec 1, so often engaged in reinforcing his roleas senior management and not as a team player, acceding toTechnician 1 and Manager 1 as they rehearse and justifyarrangement of text and its wording:AE 1 We’ve got to sell them the benefits -- why wethink it’s right. I think we should do it in[this section]T 1 There should be a general statement somewhere.M 1 That would be in the executive summary.AE 1 Yeah.T 1 And then you [a reader] would go down and findit [in the subsection] . So you’ll mention it -- very high-level --AE 1 Yeah.***T 1 We’ll have to work together on that one. We dothe bare bones technical requirement and thenwe can wrap stuff around.M 1 (writes) I’ll just call it -- “selection -- or(pause) better words?AE 1 Well, we do have “strategy”....T 1 No, it’s got to be --AE 1 Oh! Okay okay. It’s got to be -- what did youcall it?160Ideas for presentation, it seems, can come from any source.However, although Accountexec 1 accepts his role as drafter ofthe executive summary, he continues to be reluctant to take onother duties. His task avoidance unites the rest of the teamagainst him, and leads to slow-downs (schedule interference),tearing at the solidarity of group members’ interdependence(threatening team maintenance). Humour here functions toregulate the team member’s behaviour. It begins gently,maintaining the working relationship among team members andyet reprimands the task-avoiding behaviour. Accountexec 1 hasside-stepped the drafting task and proposed one for himself asa reviewer. The technicians challenge his ability even inthat role. Manager 1 must once again maintain the schedule bycutting off the bantering, which is beginning to becomesarcastic and potentially team threatening. He returns thediscussion to the task at hand:M 1 [Accountexec 1], do you want to write that?(silence. T 2 laughs) I mean that’s ourapproach.T 1 That’s key. That’s the most important thing inthis whole document.M 1 It’s either you write it or I write it.AE 1 How about if you write it and I’ll QA [qualityassure] it -- with a red pen.T 2 You know how he loves his red pen?T 1 I spent three weeks with him red-penning me(laughs).(AE 1 laughs)M 1 Did it wash of f? (laughs)161T 1 I kept putting the same document in front ofhim --AE 1 Yeah. It kept looking familiar to me!(M 1 reads next heading from the outline.)Of interest here is Technician l’s later private report to methat the document he mentioned here, after being reviewednumerous times by Accountexec 1, was eventually pronouncedsatisfactory. Technician 1 had not changed it from the firstdraft. Such events seem to threaten the team’s confidence ineach other as valuable team members all contributing to thetext production effort. It is nevertheless interesting thatTechnician 1 does not disclose this information in themeeting, perhaps in an attempt to maintain the team.Checking to see that writing styles are merged is anotheractivity of Level I collaborations. Decisions on a format,for example, can help individual team members to compose asingle document with unified presentation. The proposalbecomes a “company authored” document. The appearance of thetext on the page, for example, should be an issue relativelyeasy to decide; however, past frustrations can impede even themost straightforward collaborative task. Accountexec 1 isproviding advice to Manager 1, and Technician 2 suggests,probably unfairly, that the standard suggested is one uniqueto Accountexec 1, not one decided on by the group or preordained by the company. Accountexec 1 must call up thegeneral manager (in absentia) as final arbiter of style, inorder to defend the formatting he uses.162A more productive approach might have been to justify thepresentation approach by, for example, ease of reading by theclient, since Level I thrives on the authority of a draftingskills based collaboration. Instead, Accountexec 1 jumps to aLevel III status based argument. The possibility for alearning experience is lost. Also, Technician 2 has notallowed Accountexec 1 his role as experienced proposal writerand therefore has threatened the timely production of thetext. Debate is focused once again on the use of clip-text,in this case the use of diagrams and format taken frompreviously written documents:M 1 Are you going to use diagrams?T 1 The same ones we used before -- some minormodifications.M 1 I’m thinking about the style now.AE 1 This format (shows him a completed proposal).M 1 (flips pages) Okay. Lots of bullets --T 2 [Accountexec 1] likes bullets.M 1 And tables.T 2 [Accountexec 1] likes tables!AE 1 It’s [the general manager] -- he taught us tobe simple. Not too many words. Simplicity.Feeling pressed perhaps, Accountexec 1 retreats to theexpertise of the general manager, rather than explaining, forexample, how effective a point form document might be to theBorden audience. If he has indeed learned a presentationlesson from the GM, he does not seem able to pass therationale on to the proposal writing group.163As members leave to begin tasks on their own (though not,probably, without remembering their team members), Manager 1makes final efforts to keep each individual aware of the teameffort and the time pressures of looming deadlines. Now thatthe group is moving from a checking phase Of Level I to abuilding phase, it will be easy for them to drift apart.Manager 1 is maintaining the schedule by giving them a phasedeadline, and by providing them with a place for pieces oftext as yet another way to check on progress as well asquality of work:M 1 Is everybody comfortable that they can getsomething to word processing on Monday?T 2 Something!M 1 I’ll start a workbook.T 1 We’ll start slapping things in it -- a page ortwo and then you can start red penning enoughthat we can re-visit them.AE 1 We should include -- put in your thoughts...M 1 Notes -- lots of notes.With aid of these notes, the proposal team may continue towork together even when apart.Event J. Level I. Down To WorkA vivid example of collaborative text-building is shownby Technicians 1 and 2 immediately after Event H, as discussedabove. Of note is the speed and intensity of the conversationthey engage in. They deal quickly with difficulties of ideapresentation as they arise and concentrate on getting words164down on the page. Analysis of video-tapes reveals also thatat this point collaborators begin to eat and drink together—apples, granola bars, pizza, chocolate, pop, juice, coffee—sharing sustenance and text production ideas! HereTechnicians 1 and 2 settle in for the drafting effort. It isinteresting to note the several collaborative tasks which areproceeding at once in this brief conversation. As timebecomes shorter it seems team members must juggle more ballsin the air. The image is particularly apt in this instance.Where are the balls? How many balls must we juggle? Whatpattern are we weaving with the balls? Will our audience likethe results? What if we drop them at the crucial moment?Unlike jugglers, however, collaborative writers must also namethe balls and acknowledge the effect of the names on theentire presentation:T 1 (refers to list of headings) What else is inthere?T 2 “Information technology”? That’s what’sconfusing.T 1 I think that’s the wrong name. It should becalled “hardware architecture” or --T 2 “Network architecture” -- no --T 1 I think it’s a combination. “Network andPlatform”. I’m treating it like that. That’swhat I want to cover of f with [Accountexec 1]***T 1 We can extract this from our [previous report]T 2 Are we going to be able to get all this to wordpro for Monday?T 1 It’s going to be tight.165T 2 Some cut and paste --T 1 We won’t have ... (lists several missing piecesof information).There is some evidence here as well that the solution to thetechnical problem is inextricably interwoven with the buildingof the text in Technician l’s comment that he is building thenetwork and platform together as the solution, and not merelydiscussing them together in the same text section. Once againhere as well is the concept of real world writing as “clip-text”, of writers assembling the already-written in the textproduction effort.Day SevenCollaborative writing on Day Seven results in thecompletion of network management and a description of theNovell network by Technicians 1 and 2. Manager 1 is alsodrafting on his own.Event K. Level I. Covering Of f andEvent L. Level II. The Audience AwaitsTechnician 1 and Accountexec 1 meet to further discussthe presentation of ideas. Technician 1 is “covering of f,” orchecking for approval of his approach, both technically andrhetorically, with Accountexec 1. Although Accountexec 1 iscarefully avoiding any contact with paper—which remains inthe hands of Technician 1 at all times and thus, perhaps,remains fully his responsibility—Accountexec 1 and Technician1661 work together amicably for once, and resolve an approach forthe drafted copy of the proposal:AE 1 You’ve got to keep it sort of focused -- (makesa square with his hands in two choppingmotions) otherwise you can spend a lot oftime...T 1 Well, that’s what I’m saying. So we’ll keepthis sort of low level.AE 1 I would not have any more than four optionsthat could be added.T 1 That makes sense -- two for each.Technical and rhetorical issues are closely linked here onceagain. These collaborators agree that keeping the technicaloptions limited will also limit the complexity of the textformat. Also, as in many Level I collaborations, work isaccomplished quickly between two motivated people.Once this brief conversation on drafting is completed,however, the level shifts to Level II, and another meetingseems to begin. There is a new topic—dealing with thevarious audiences of the Borden proposal. An effective teameffort would have drawn the vendor partners (one audience) andclient (another audience, or perhaps audiences, Jake and Mr.B.) into a sympathetic joining with Cerebellum. Variousfactors, including the general proposal strategy of testingfor the best partner, seem to be hampering team maintenanceand audience sensitivity, thus making text production effortsmore difficult. In response to Accountexec l’s queryconcerning vendor responses, Technician 1 reports: “Vendors?They’re not happy with us .... They’re not too impressed. And167also they’re not too co-operative on pricing.” Obviously, notonly are relationships being strained by the general strategywhich favours one vendor, but Technician l’s informationgathering efforts have been severely hampered as aconsequence. Vendors no longer feel as though they are teammembers, and furthermore have ceased to be even co-operativeoutsiders. They strike back at Cerebellum by withdrawingaccess to information, thus obstructing the proposal team’stext production in general and hampering especially schedulemaintenance efforts. Missing vendor information will continueto hinder the proposal building effort from this pointforward.The adversarial roles of audience and speaker continuewith the Cerebellum team’s efforts to deal with the Bordenproject manager, Jake, and Borden’s owner, Mr. B. Manager 1enters the Event K meeting above between Technician 1 andAccountexec 1 who, ironically, have just been discussing howto deal with formerly sympathetic vendors turned uncooperative. They must now deal with the ongoing problem of aproject manager (Jake, first mentioned in Event A) with whomno one has developed rapport, and a rather distant clientowner, Mr. B., who seems to be the real decision-maker.Manager 1 has engaged Jake on the phone in what could be aneffort to include him as a team member, and reports hisconversation to Accountexec 1 and Technician 1. How does onedeal with a difficult audience? In this case, team membersseem to agree that the project manager is foolishly keeping168vital information from them, thus hampering the proposaleffort which is after all intended to solve Borden’s problems:AE 1 And he [Jake] was very open with you?M 1 Yeah, and in the end we kind of closed. I said,“Is there anything we’ve missed, or anythingyou think is important in this approach,anything you still have a problem with? And hesaid, “No, we’ll weigh it against thecriteria.”AE 1 Yeah. He doesn’t have the selection criteriain the proposal, does he?Ml No.T 1 No.M 1 Which is a mistake. How can people bid ifor write a proposal if --T 1 They’ll go with their best knowledge.Although such agreement builds the cohesion of the Cerebellumteam, it does so in a rather unpleasant manner. Cerebellummembers do not invite the audience in as an equal team member.Instead, they close ranks and maintain themselves as a groupby turning the audience into a target. Having agreed onceagain that Jake can be safely ignored, perhaps because he isdifficult to relate to, or more likely because he has now beenjudged by the group as less important than the rather shadowyMr. B., the team works out the owner’s probable criteria as amember of a group they call “senior executives”:M 1 So I think their selection, for [Jake] becausehe’s the delivery guy, are times, limits andcosts.AE 1 Yeah. In that order. (laughs)169M 1 ‘Cause he’s the delivery guy. If you asked[Mr. B.] what his --T 1 Business related --M 1 Would be costs and flexibility.T 1 He can plan -- he can budget --M 1 He can stretch things Out. And my experience -- I don’t know about you guys -- with seniorexecutives is they don’t want to make hugedecisions. They want to make little decisions.AE 1 Yes. Absolutely.Echoes of Aristotelean advice reverberate here. Members ofgroups share common behaviours and likely reactions torhetorical ploys: little decisions on options should win overMr. B as the primary audience.The Sundial proposal also faces problems with deciding ona presentation which will convince the client-audience. Incommon with all potential clients, such audiences are powerfuldecision makers who are often weak in technical knowledge.Especially for technical specialists, finding the appropriatestance towards the client which can be embodied in theproposal, often proves to be difficult. In the Sundialproposal effort, however, Technician 7 proves to have theability to relate to an audience. Here is a conversationwhich seems to be on the cusp of a traditional adversarialview of audience and a competing sense of audience as a memberof the team. Technician 7 shows an understanding of theclient’s confusion over technical matters, and a willingnessto work with them to draw them onto the technical team by170giving them knowledge and a new way of defining terms. Hisconception of the audience as new to the implications oftechnology leads him to an interpretation of his role asfacilitator, one who would help them to become informed. Thereaction of the Director shows immediate suspicion--a view,perhaps, of the audience as capricious, powerful, andpetulant. Such an audience, in the more traditional view,must be carefully handled, kept at arm’s length, and watchedin case they become annoyed:T 7 They’ve said through these meetings: “Don’ttouch the workflow, just leave it as it is.”I’ve said, “Fine, I hear you. I’m going toproduce a workf low and a modified workf low.Why? The technology changes the workf low.” Theother vendors are avoiding that, theconversation of workf low, because they get shotdown. “Don’t touch that, don’t touch theworkf low.” We’re going to touch it.D 4 Why?T 7 Because when I sat down and asked them thequestion: “You recognize the technology isgoing to change the way you do business,”didn’t use the word workf low, “that means someof the steps in your workf low are going to bedifferent.” “Oh, yeah, yeah.” “What are thosechanges?” “We don’t know. We’re looking at thevendor to tell us.”D 4 They said that.T 7 Exactly.D 4 Who said that? Someone with authority?T 7 [Allan T.j Guy that owns the thing.***171D 4 Why did they say “don’t touch the workf low”then?T 7 It’s a definition-syntax thing. Okay? Theydon’t want you to go in and do a re-engineeringeffort which is what the people [some othervendors] want to do. They also want to make itclear that there isn’t a clear linear flow incentral records, there isn’t .... I said, “Whatare you trying to achieve there?” He said,“What we’re trying to achieve is work teams:” Isaid, “You’re not there yet.” He said, “No”. Isaid, “Are you looking at the imaging solutionto facilitate this?” He said, “No -- could itdo that?” And that’s where I see our vision asunique, I don’t think the others have caughtthat.AE 3 This is where the vision starts to come in.Let’s have some more. All right, I like yours.D 4 We just have to present that very carefully soit doesn’t sound like we’re throwing out theirassumptions.The technician persuades the others that his bold and riskyplan can work, if presented carefully. He is able to modifyhis view to include the director’s concern and suggest a wayto word the “vision” to win the audience’s trust and educatethem about their needs. Technician 7 is a collaboratorendeavouring to be sensitive to two audiences, both the clientand the Cerebellum team.Event M. Level I. A Decisive DyadDespite Director l’s assertion (see Event A) that theBorden solution should be complete before the writingproceeds, the text and solution building proceedconcomitantly. Here we see Manager 1 engaging Accountexec 1172in a discussion of the solution which must be clarified.Although Accountexec 1 initially moves physically away fromthe problem, and continues to avoid it through language,Manager 1 presses him sharply. Together they do eventuallygrapple with the search for a completely detailed andrecordable solution, coming back to the task of rehearsal, ofdeciding which sorts of words fit:M 1 This is where it gets a bit fuzzy in my mind.How are we going to do this?AE 1 (moves back from looking at M l’s computerscreen) I don’t know.M 1 Do you have any ideas?AE1No.M 1 Does [Technician 1] have any ideas? We’regoing to be asked this.AE 1 Yeah.14 1 “How are you going to do this?”AE 1 I don’t know. Well, I presume that what wehave to do is -- uh -- set up some sort of test(outlines idea)M 1 ... so maybe “Our ideas are flexible but oneidea might be . . .“AE 1 Yeah .... That’s possible right? So we couldhave multiple levels, right?Accountexec 1 seems enthusiastic at last about acceptingManager 1 as an equal team member. Skill—based authority liesequally with both team members in this instance. Equalpartnerships are relatively rare in the Borden proposaleffort, with the notable exception of Technicians 1 and 2 aspreviously discussed (for example, during Event H)173During the remaining portion of this event, Manager 1 andAccountexec 1 continue to maintain the team, working quicklyto resolve differences, include ideas from absent team membersand build the text as they develop the solution:M 1 Director 2 was suggesting this kind of chart(draws).AE 1 Yes yes -- .... that could be on the secondpage -- then we could have (points to page,then draws). How does the [project) team mesh?Z4 1 Yeah. It meshes this way (draws). Right?AE 1 So, we’ve said “This is the whole project --this is yours .. . .“ But his [Director 2’s]thinking and mine [differ] (draws difference)but we can massage that later on.M 1 But what happens is that here (draws) you’retogether. So we’ll call it “AcceptanceCriteria” -- it’s less threatening to him[Borden]AE 1 Right. A selection. Absolutely. I like it.Charting is writing, or composing, too, in a veryimportant way. Building the chart together functions inseveral ways. First, the chart will form part of the finalproposal package; therefore, drafting is accomplished andadvances the process, keeping the team on schedule. Second,charting allows writers to discover and merge different ideason the approach which will be presented. It provides aconcrete referent for collaborators to demonstrate theirthinking and evaluate ideas brought by fellow collaborators.Finally, it serves to remind them of the broader audience theymust address. Director 2’s ideas must be recognized, as must174the reaction of the client. Given earlier noted problems withaudience, it is not surprising that Borden remains somewhatnebulous here—is the hlhimu mentioned Jake or Mr. B? Thecollaborators seem content that it could be assumed that noaudience likes to feel threatened and the chosen wording istherefore safe. Notable too is the fact that the visual hereis original text, unlike clip-text art referred to previouslyby Technicians 1 and 2 (Event H).Day EightDay Eight marks the first and second drafts of manysections of the technical solutions, written by Technicians 1and 2 and Manager 1, and thus the proposal can be said to bein a first version. For example, organization of personneland the work plan for the project are in a rough draft. Theproject team members’ descriptions are begun, with twopositions outlined. In addition, Accountexec 1 has completeda draft of the executive summary.Event N. Level II. Collabo-techTechnician 1 continues to gather information, checkingfor problems which have been raised during his involvement intext drafting events. He realizes that, as technicalarchitect, he will bear much of the responsibility for thetesting process which forms the heart of Cerebellum’s proposedapproach. For assistance, he seeks out another technicianwith more specialized knowledge than he has. Worthy of note175is that Technician 4 is two steps lower on the officehierarchy; however, he holds the authority of knowledge duringthe following conversation:T 1 (describes strategy) How much effort would werequire? Two weeks?T 4 You know who’s the best person to talk to(searching for information as well in a tradejournal as he speaks)? [Barb] ... to do itproperly, they’ve budgeted two to three months.T 1 We don’t have that. We have max two weeks andwe don’t want to spend a lot of money doing it.***T 4 Well, you’ll have to simulate the process (T 1records) . Design is two weeks in itself(discussion continues)Ever present time pressures necessitate a course of actiondifferent from the one proposed by Technician 4 as a model.He listens to new information and suggests an a1ternativwhich might meet Technician l’s criteria. The Level IIactivities of information gathering and considering continuedespite the now fairly advanced state of the document andgeneral predominance of Level I activities.Event P. Level I. Enter the Dragons, Part AThe RFP document both initiates and sustains the proposalwriting process. It lays out the requirements and even theformat for the proposal, and as such should be consultedconstantly and frequently by the writers. If regarded as yetanother collaborator or member of the team, it could serve asa chief facilitator in text production. Like all such176documents, however, the RFP is open to interpretation. Often,alternate interpretations are valid, even after carefulreading, and must be discussed by the team to ensure a commoninterpretation. At other times, however, a lack of closereading leads to problems which could have been avoided byinvesting extra time in analyzing the RFP text. The proposalwriting process is also complicated here by the presence ofother documents prepared for Borden by Technicians 1 and 2which have been used by Borden in the preparation of the RFP.Here, for example, we see Manager 1 and Technicians 1 and2 struggling with the RFP and other documents in theirattempts to draft their own text. The proposal manager wantsto maintain a simple and straightforward approach: the RFPasks for this; therefore here is our response, in a one to onecorrespondence. The vojce or ethos projected by the RFPbecomes, for him, a clear audience. For thetechnicians/engineers, however, no such clarity exists.Although they allow that the RFP asks for certain specifics,they see, perhaps because of greater technical knowledge, thatthe answer is not straightforward. There is rather acontinuum of answers, all of which satisfy to some degree.The answers, moreover, are not completely pre-packaged butare, to some extent at least, chunked (and available as cliptext) . They have been previously rehearsed in otherproposals, and chosen from among a repertoire of technicalpossibilities and all of which come with hefty price tags:177M 1 Does each option relate back to a requirementin the RFP?T 2 Well, there’s different options to the degreeyou want to go and satisfy that requirement.H 1 Yeah. I think we’ve got to be -- I don’t thinkthey’re looking for a study. I think whatthey’re looking for is a proposal***T 1 We’ll make it clear to them that these are allthe possibilities -- and they’ll use thatinformation to make that decision. So if theyneed this [option) -- they’ll go to thatsectionH 1 I think if we can tie it back to hisrequirements in the RFP - -T 1 He was very vague .... He wants totalavailability [of the computer system] -- 99.8percent [of the time)T 2 I don’t think he realizes what it’s going totake -- the dollar figureT 1 It’s a busy little section. So -- we just haveto round it out, polish it off and make itclear in the document.H 1 Where is the document?T 2 The RFP? I’ve got it here.H 1 You see the way I was looking at the RFP wasthese requirements are outside the base system(he points to relevant section, and all threemove in close to read)Text drafting here is also complicated by the presence ofother documents prepared for Borden by Technicians 1 and 2which have been used by Borden in the preparation of the RFP,and by the technicians’ prior experience working with theaudience (probably Jake, given the rather pejorative tone oftheir comments) who is regarded with some suspicion. All of178these factors (technical knowledge, previous documentation,and previous audience experience) lead to audience sidestepping. The technicians will not ignore the RFP or Jake asits probable author, but instead attempt to transfer theirresponsibility to propose a solution by listing possiblesolutions and asking the audience to pick their own. Thus,perhaps, they try to avoid writing persuasively. Manager 1,however, continues to insist that the RFP be more closelyscanned for text drafting ideas.Although Manager l’s lack of experience with the audiencemay be leading him to rely too heavily on the RFP document, hedoes remind the team of important audience concerns. Thetechnicians, over-reacting perhaps to their previous somewhatnegative experiences with Jake, agree to reconsider the RFPdocument. All collaborators then discover that they have notbeen reading material closely enough:T 2 (reading aloud from RFP, then -- ) Can theapplication be split like that?T 1 No.T 2 I didn’t think it couldT 1 It could be in the future [rumoured updates notyet available]T 2 (ignoring T 1, pointing to RFP) He’s gone andthrown all these things back in -- we wentthrough this [with him] and told him [whichones could not be done] .M 1 okay, what about this next thing?T 1 Those are the words. We have a table [whichshows the idea] in the smaller document --179M 1 I missed that.(T 2 searches for and points out diagram)Lack of close reading has lead to difficulties in draftingtext. Schedule maintenance is also threatened by continuingproblems in casting the audience in some helpful fashion(rather than as an obstacle) . We see here also the continuingimportance of visuals in the drafting effort, as ways toencapsulate the technical solution. Nevertheless, “thewords,” as these collaborators frequently say, must bewritten. The drafting process, however, has been aided by theclose reading of relevant documents.Merging of text to achieve a single version with a singlestyle, or corporate “voice,” is yet another collaborative taskconsidered during Level I. Although one writer willeventually resolve differences in presentation, the group mustrecognize the problem and minimize the time it will take tocorrect The maintenance of Level I roles within the teamcontinues, it seems, to be a problem here. Both Accountexec 1and Technician 1 are unhappy with drafting duties, and indeedthere seems to be no commonly accepted understanding of whomust accept drafting activities, other than the proposalmanager. Interestingly, Accountexec 1 has written one summarybefore the proposal document is substantively complete, andyet uses lack of “meat” in the rest of the proposal to excusehis lack of progress in his other summarizing duties. Suchtask-delaying impedes the progress of text production or atleast points to an unusual conception of a summary. The group180has, however, agreed that a draft must be completed beforestyles can be merged:(AE 1 enters room and remains standing)M 1 Is that your [executive] summary?AE 1 Yeah.T 2 Thank you.M 1 They’ve just been fixing mine [the text he hasdrafted] -- told me to take out my lies.AE 1 So you guys can take a look at it. It’s rough.The concepts are there.T 2 The system objectives?AE 1 Pardon?T 2 You did the system objectives?AE 1 Uh, no. I want to make sure this is okay first.And then once I have some meat from your stuff,I can summarize it in the system objectives(starts to leave)1’! 1 When are you coming back for changes?(general laughter)AE 1 Whenever -- I’ll be in my office.T 1 Maflana.(AE 1 leaves)M 1 (reading AE l’s text) Two different styles.T 2 No -- we’ve got about four different writingstyles!T 1 What we were thinking was -- we’ve got to getit to a level where it works and then integratethem. Otherwise, it’s not going to read verywell. Once everybody’s happy, then I think wejust have to do it in one style. And it’s notme!(T 2 laughs)181More collaborative reviewing and editing might aid thisprocess. However, in a document which inevitably involvesseveral writers (and pieces of clip-text lifted from otherdocuments written by still other writers), style merging, orthe formulating of a single company voice, will always be aconcern.Mid-process activities may thus be seen as intense workin Level I activities—building and checking text forpresentation in the proposal. The middle phase also includesLevel II information gathering and considering which furthersthe Level I drafting of text. Several collaborative textproduction principles give this real world writing itsdistinctiveness. These principles include the use of cliptext, the importance of visuals, and the constant merging ofstrategies and tactics to achieve a unified and consensualapproach to the writing of the proposal.182CHAPTER SEVEN: PRESENTATION A2D DISCUSSION OF LATE EVENTS“Now! Now!” cried the Queen. “Faster! Faster!”And they went so fast that at last they seemed toskim through the air, hardly touching the groundwith their feet, till suddenly, just as Alice wasgetting quite exhausted, they stopped, and she foundherself sitting on the ground, breathless and giddy.(Lewis Carroll, Through the Looking Glass, and whatAlice found there)Time pressures and other accumulating problems (such asthe lack of role maintenance as leader and participant byAccountexec 1 and lack of audience focusing by allcollaborators) leads to a sudden increase in the involvementof senior management in the proposal effort during Day Eight.As Alice finds in her dealings with the Queen, being involvedin a sudden decisive act by a senior player can provedisconcerting. In the collaborative text production processof the Borden proposal, the time, it seems, has come toaccelerate the pace. Authority begins to rest with highstatus Level III personnel. Directors 1 and 2 take prominentleadership roles over all collaborations.Event P. Level I. Enter the Dragons, Part B.As Event P on Day Eight continues, we see Director 2enter the meeting in his role as senior technical director,and begin to collaboratively resolve problems over draftingissues earlier discussed by Manager 1 with Technicians 1 and2. At this point in Event P, Manager 1 has been diagrammingon the white board, sketching approaches to the presentation183of the solution in the proposal and using a visual to shareand check his understanding with the group. Due largely tothe explanation of Director 2 during the sketching activity,Manager 1 finally realizes why the RFP requirements cannot bemore simply answered in the proposal:M 1 Ah -- so we’ve got base systems that look likethis (he draws a very small square and a verylarge square)D 2 Yes!T 1 Some are not even square -- they could beround!(general laughter)***T 1 Past a certain point -- there’s nothing similarbetween those machines. And how you manipulateit becomes really difficult. So it’s not asstraightforward - -Visual approaches, to both sharing and checking understanding,seem to be one effective way for collaborators to discussideas. The group’s affinity for diagrams of various sorts, assupplementary text, continues to be a theme in this meeting.Director 2, who now emerges forcefully as the leader of thisLevel I drafting session, invests a great deal of confidencein charts and tables as effective textual devices, which canserve as guides for the drafting of paragraphs. Time isclearly growing short, and Director 2 tries to deal withschedule maintenance by taking control of the group’scollaborative drafting process. He and Technician 1 have beenworking together on a spreadsheet, obviously without Managerl’s knowledge:184D 2 I wouldn’t worry in the words as to what ourrecommendation or approach is. I think we’vegot it fairly clear.M 1 I know the approach. I’m just trying toorganize it.D 2 We don’t want too many choices for them tomake. “If you want the absolute base, it’sthis.” The words aren’t difficult and thespreadsheet.M 1 (not hearing last few words) Can we get them aspreadsheet that shows that --D 2 I’ve got it all made up ready to go.M I. With the technical requirements --D 2 I’ve got it all made up ready to go. That’sour problem -- we need the numbers. I justphoned --14 1 Can I see a copy [of the spreadsheet] --D 2 Umm, hmm - - so if we write the words accordingto what our schedules and spreadsheets looklike, it’s not that difficult. (T 1 sighs)Right?T 1 For the words I can just put down little blobsof information right? (laughs)Role maintenance and team maintenance are clearly at issuehere. In a Level I drafting session, an employee’s skill inpresentation usually has precedence over status and technicalknowledge. However, lack of technical understanding hashampered Manager l’s ability to draft at this point, whichshows the importance of collaborators having a broad base ofskills at all Levels. As effective collaborators, Technicians1 and 2 and Director 2 have helped Manager 1 to overcome theproblem. However, they have also avoided collaborating withhim by developing a new textual presentational device without185informing him or asking him to participate. Director 2 hasthe status and the technical knowledge to effect such achange, and is prepared to impose presentational style (whichis not one of his strengths) in the interests of meeting theproposal deadline. The team must now survive by acceptingthis sudden change in roles.As leader, Director 2 is forcefully taking over Level IIIconcerns with maintenance of the strategy, Level II concernswith information gathering, and Level I concerns with therhetoric of the final text. Other team members are lesssatisfied with their roles. Technician l’s comment (“littleblobs of information”) indicates perhaps that he is happierdealing with Level II information gathering than drafting andgeneral problems of presentation at Level I. And as themeeting progresses, Manager 1 still seems to be overwhelmedwith his drafting role, despite further reassurances fromDirector 2. As an attempt to check Level I drafting issues(the original intent of this meeting), the conversation whichfollows demonstrates collaborative tasks in conflict,threatening the team relationships.Director 2 seems to be attempting to relieve Manager 1 ofa role he cannot handle—that of technical expert, encouraginghim to record the broad strokes of the strategy and leave thedetails of the technology to other text producers. His advicehowever, is somewhat contradictory. Whereas previously (asnoted above), Director 2 seemed to be pushing the primacy ofthe visual, which would be built and then reiterated by the186words, here he seems to indicate that the words should comefirst and then the visuals will inscribe them in a visualform:M 1 Use the technical requirements in the RFP,right? There’s a base and options. (sighs)D 2 Don’t wor -- to me, it’s really important toget the words down that describe what ourapproach is, in terms they can understand, andwe’ll make sure the schedule supports thosewords . . . . “We recommend this is what you do” or“we do, collectively” but from a wordsperspective, as long as we know what they are,we can make the numbers match them.M 1 I just want to tie it back to the [RFPIstatement of need.D 2 Fine. Great idea. So what’s next?The breakdown of collaboration may lie with the suddenchange of roles at this point. Level III, Director 2’s usualcollaborative site, is concerned with the general strategywhich must be made clear in the proposal and which mayreasonably be expected to be drafted by the proposal managerand then further reinforced by diagrams. Technicalinformation, however, might most easily be presented in achart and then supported by surrounding text. Director 2 isnot as skilful at Level I, however, as he is at Levels III andII. He forces this drafting issue not by citing rhetoricaladvantage, but rather by level jumping. His status andgenerally extroverted free-wheeling style, so appropriate to aLevel III meeting, and his considerable knowledge ofinformation sources and pragmatic need to get the job done,appropriate to Level II, are used here to mask the lack of187expertise at drafting, which is a Level I issue. He is notflexible enough to engage with the text production activitiesnecessary during drafting, such as rehearsing ideas, workingintensely with questions of organization, for example.Manager l’s repeated attempts to engage collaborators in aclose reading of the RFP are quickly dismissed again byDirector 2’s concerns about schedule maintenance: “What’snext?”Day NineDuring Day Nine, the technical staff works on completingthe first drafts of the solution, now called the “TechnicalStrategy.” Most of this work is done by Technicians 1 and 2.Their drafts are reviewed by Director 2, who also adds totheir work. He and Technician 2 review results and make.further modifications together. Director 2 also does a firstdraft of the costs of the project which will be needed laterby a Level III group to decide on prices. Meanwhile, theproposal manager writes a new executive summary since the onedrafted by Accountexec 1 has now been judged to beinappropriate by the team. Manager l’s work becomes the newfirst draft of the executive summary which will move on toDirector 1 for further review.Event Q. Level I. The RecyclersText building continues apace with the technicians stillworking closely in concert at drafting the solution and188simultaneously building the text. Consider the difficultreality of their situation. They are trying to produce adocument with insufficient information. However, they arelargely responsible for gathering that information, as well asbuilding the technical solution and shaping the text whichpersuasively presents that solution. They turn once again tothe use of clip-text from previous reports they have written,and from letters composed by vendors to assist them. Thematerial they are attempting to build consists of both wordsand visuals, and they assemble it in much the same fashion asthey usually gather information:T 2 Okay, so this is generic (reads aloud, then)Where’s that --T 1 What are you looking for?T 2 That diagram you just showed me. Where’s thewording around that?(T 1 locates previously written text on the desk)T 2 Okay. (scans page) Not going into thesespecifics.T 1 It starts with recommendations ... (points outparts) and that’s just a “come-back” ... Maybe,what I have that’s better are those threeletters I got vendors to send me [on anotheroccasion] . That’s not a bad source ofinformation. (he leaves and returns)(T 2 begins copying text into proposal)T 2 I’m just putting this as an overview --T 1 And then this (points) gets merged in there?T 2 Right. (she reads aloud as she types)T 1 Where did [Director 2] get this amount [ofmoney]?189T 2 (writing on the hard copies and eating an apple)Does it add up?T 1 All the numbers? No, but there are a lotmissing. (writes on his copy as T 2 continuesto type)Some agitation seems inevitable. Technician 2 istranscribing from one document to another. She andTechnician 1 are assembling bits of information whichthey hope will need little if any re-writing. However,such pre-composed text must be inserted with some skill.Writers must pay close attention to headings and to thegeneral cohesion of logic and style. Such behaviour,while not guaranteed to produce an effective document(see their earlier comments concerning Borden’s lack ofability in including previously written text in the RFP),does nevertheless save time, time which Technicians 1 and2 need in order to keep pursuing missing information. Inaddition, because one source document, which theythemselves wrote, is familiar to the client, it likelywill be easier to integrate into the proposal document.Writing here is most obviously a “clip-art.”Event R. Level I. A Woman’s PlaceAt this juncture, the technical staff involved inthe Borden project (Director 2, and Technicians 1 and 2)are continuing to experience difficulties with thepresentation of the technical solution. The proposal iscentred on a suggestion that three different hardware190platforms be tested in order to ensure that the clientwill get a satisfactory product to meet their computerneeds. As struggling Level I drafters, the technicalcollaborators seem to realize that the words areimportant, that the phrases they use to explain theirsolution will have a significant effect on the client andare worth discussing. However, an authority issue arisesin the debate, one which most significantly involves thefemale technician.These collaborators are all more experienced withLevel II collaborations, such as gathering andconsidering the impact of information. We see them herein a Level I collaboration. Director 2 asks forconfirmation on an appropriate heading from Technicians 1and 2. Whereas Technician 2 is almost always in asituation where she lacks authority (being lessknowledgeable and having less status), as a drafter, sheseems to feel more on an equal footing. Here, sheattempts to exert her authority by challenging all othersat the meeting with their choice of the word “trial”:D 2 One thing we do have to spend time thinkingthrough are all the things under this “proofof” -- I don’t know if I want to call it “proofof concept”, or “proof of practice” or “proofof --“ something, but that’s what we’re talkingabout. It’s “proving in” the technology.“Prototype”? No.M 1 “Trial”.D 2 “Trial”. Yeah! “Hardware application trial”.(to T 1 and 2) Is that what you’d call it?191T 1 Yeah. “Trial” is good.D 2 Well shit. We’re not doing “proof of concept”.The system’s in production.T 1 We’re just putting all the new technologytogether.T 2 We’ve moved from “benchmark” to “proof ofconcept” --M 1 “Prototype” -- “trial” --D 2 I don’t want to scare them that we’re doinganything significantly leading edge -- “proofof concept” suggests something that’s neverbeen done before, but this has all been donebefore. Different packaging. So we’re justtrial-ing the environment and the application.T 2 “Trial?”T 1 “Trial.” That’s good.D 2 Fine. We have to talk about what the activityentails.T 2 I don’t know if I like “trial”.D 2 Which?T 2 I don’t know if I like “trial”. It sounds asif we’re trying something --D 2 Director 1 will come up with a different goodsolid word.T 1 Oh yes, oh yes. Well, we’ve got the basesystem, so do we want to identify(continues discussing testing procedure)Her attempt at disagreement in this instance is shortlived, however. The ultimate drafter, Director 1, iswaiting in the wings to solve these problems, according toDirector 2. The meeting participants return to discussingthe details of the technical solution, a more comfortabletopic for most of these participants and one thatmaintains their work on Level I drafting.192Later in the meeting, however, Technician 2 objectsto wording once again, this time with quite differentresults. Technician 2 establishes her authority in thisdebate over a properly worded heading by using technicalknowledge which she has, and Director 2 obviously doesnot. Here, however, she seems to prevail partly becauseshe establishes her point of view by using a technicalexplanation to support (or rather not support) a choice oftechnical terms:0 2 That’s what I call “LAN expansion”.T 1 That’s key stuff.M 1 Is that a third option besides --T 2 I don’t know if I’d call that “LAN expansion”.D 2 What would you call it?T 2 Because they’re still going to be going overjust the dial-up lines ... (explains, over topof T l’s supporting explanation)O 2 (maintains eye contact with T 2) Oh oh oh ohoh. So it’s not “LAN expansion”. (writing)That’s part of “Fault Tolerance”. (looks at T2) “System Fault Tolerance?”T 2 Yeah.Such knowledge is more likely to gain respect with a groupof technical experts than her attempts to protest the word“trial” based mainly on her sense that the group did notwant to be seen as merely “trying” to find a solution.Director 2 reinforces Technician 2’s role as a valuableteam member by validating her explanation, and asking forher approval of a different heading. With theseconversations, Technician 2 has established more authority193as a valued collaborator in both Level II and Level I.Director 2, in fact, will later invite her (see Event V)to be a reviewer of text which he has written. Technician2 has indeed found a place for herself on the team.Event S. Level II. Pulling Rabbits From HatsOne of the most challenging collaborative activities ofLevel II is the estimating of personnel and time which will beneeded to put the proposal into effect once it is accepted andbecomes a project. Level III pricing quotations cannot beformulated until the costs of the project, including the wagesof employees, are factored into the proposal pricing. Goodestimators are a rare breed. They understand the technicaland business solution intimately, know the expertise and timeneeded to put the solution into practice, and are able tobalance complex issues (such as employees who might handledifferent jobs as long as those duties do not overlap). Otherfactors, such as travel time and flexibility to deal withunexpected contingencies must also be worked into theequation. In the end, however, good estimators work from afelt sense of rightness, based on experience and generalexpertise, ratherthan on equations or rules.Here we see Director 2 enjoying a demonstration of hisexpert knowledge as an estimator. The marked contrast betweenTechnician 1 and Manager 1 in their responses to this ratherremarkable display of estimating may be due to their differentviews of their own roles as team members. Manager 1 has194initially looked for a consensus, or at least averaging amongseveral estimating worksheets. Clearly, however, Technicians1 and 2, with greater experience in working with Director 2,are prepared to sit back and watch, conscious of theirrelative lack of knowledge in estimating. They bow, as LevelII collaborators often do, to the one with the most knowledge,and prepare for roles as reviewers of numbers generated.Director 2’s demonstration seems to result in some difficultyfor Manager 1 as a proposal leader. Of special note as wellis that the building of text here is both rapid and original.Director 2 does not rehearse or seek other text to support hiswork. Numbers may be handled differently (much the sameactivity will be done later involving the development ofpricing quotations):D 2 This is how we do estimating, right? Pull anumber out of the air. Comfort level. (hestarts to record figures on a worksheet.Manager 1, Technicians 1 and 2 watch, althoughthey each have a worksheet of their own.Minutes pass.)Is everyone doing their own estimate?MlT2 No.T 1 No. I’ll argue any number though.(Minutes pass. D 2 still scribbling)D 2 (looks up) Who’s doing user training?M 1 (D 2 already has head back down) I had an “x”under us for doing some --D 2 There we go!M 1 Are you a fast estimator?D 2 Yup! Always have been.M 1 Fast and accurate, right?195D 2 What was funny was -- we had this other guyhere and we’d read the RFP, flip through it, goaway. And [come back and) say “Here’s ourestimate!” and inevitably we’d be less thanfive percent apart. No dialogue. Nodiscussion. Let’s go! (totals numbers, sottovoce)T 1 Amazing stuff.D 2 137! Divided by seven [months available] is --T 1 Three days off! [assuming twenty-two days in amonth]M 1 Don’t forget we have some overlap --D 2 (hands sheet to M 1) Do I pass?!M 1 I have to put these against the time line --D 2 Oh sure!M 1 But for now I guess --D 2 I mean I was not thinking seven months when Idid that -- but it comes out to seven months!Director 2’s ability has, it seems, been developed with thecollaboration of another employee. Together they learned howto estimate and helped to shape each other’s work by comparingnumbers. We do not see the struggle here which Director 2faces in composing words, just straightforward invention ofnumbers which can be used as information in the proposal.Day TenDay Ten marks the production of the first total proposaldraft which reflects a complete re-organization of theproposal according to the headings and subheadings as definedby the RFP. Although much of the re-ordering is fairlystraightforward, some sections are new and must now be drafted196from the beginning. Assembling different versions of textalso reveals that some sections have been done twice bydifferent writers. Many sections have been drafted but notyet reviewed. By the end of the day, three sections stillneed a first draft: the covering letter, pricing, and theappendices. The text has become a document of some twenty-six pages. In addition, by the end of Day Ten, the style ofthe proposal has also been modified; the proposal will berewritten to include more “sell.” The draft now enters itsfinal version.Event T. Level 111/I. Asking the Tough QuestionsAs time grows ever shorter on the second to last daybefore deadline day, we see one more senior employee, Director1, step in to try to maintain the schedule. Next to thegeneral manager, Director 1 has the most status in the firmand is also the most respected writer. This makes him aformidable presence as both a Level III and Level Icollaborator. We see him here for the first time since EventA. As a Level III collaborator, he uses his authority as anemployee of very high status to challenge the now questionablerole maintenance of Accountexec 1. As a Level I drafter,Director 1 must also understand the state of the document sofar. He institutes a text-checking procedure, which quicklybecomes a check on team maintenance. This is obviously adifficult situation. Director 1 asks brusquely for a reportof Accountexec l’s role maintenance behaviour on the team.197Manager 1, responding to his higher status, reluctantly tellshim the problem, which it seems, Director 1 knows already.Team maintenance now must give way to schedule maintenance:D 1 [Go through this] just so I get a feel for howthings are.M 1 Right. (points to list of sections) This needsa re-write. Don’t have that.D 1 And [Accountexec li’s doing that?M 1 Well, yeah, he was going to do that.D 1 How much is [Accountexec 1] in this process?Ml Urn-D 1 In and out?M 1 He’s basically not in the last day or two -- atall.D 1 Does he need to be?M 1 Um -- I’m not sure how much value added -- nowwith Director 2 --D 1 Uh -- okay. Time might be all. If we need himto -- because Director 2’s not going to get toit --M I Uh -- my experience [says?] re-write here --D 1 So, who’s got this (pointing to a section oftext)?M 1 Uh, leave that. Let me check with [Accountexec1]D 1 Even if it doesn’t have the best structure,just the ideas on a piece of paperCollaboration, it seems, struggles—especially when teammembers do not maintain their roles. This will more certainlyhappen when they are not satisfied with what those roles are,or do not fully understand what they must do to fulfill thoseroles. The progress of the text will suffer because sources198of text portions, such as those which could be provided byAccountexec 1, are no longer available. Director 1 musttherefore move to another approach, shifting Accountexec 1into a new role as a reviewer and finding someone else toprovide the text needed.Trying to change the roles of individual team members atthe eleventh hour seems a generally inadvisable tactic.However, precisely because of the closeness of the deadline,this is what happens. Because Accountexec 1 has notmaintained his role as team leader and drafter, the group isnow in danger of not maintaining their schedule. Director 1,earlier in the event, has suggested that even though Director2 has picked much up of the responsibility as proposal leader,Accountexec 1 should still be composing some portion of thedocument. Now however, Director 1 seems to feel that othersshould be drafting and Accountexec l’s role should bereviewing and maintaining audience focus. Manager 1 has beenput in the uncomfortable position of breaking team solidarityonce again. He must also defend his own attempts atresponding to Accountexec l’s lack of contribution to thedrafting of the proposal. Director 1 forces the issuesomewhat brusquely, no doubt annoyed at the difficult positionthe proposal team now faces. Understandable or not, however,Director l’s actions strain team relationships. However, healso points out an important issue in writing-as-clip-art:the provider of text must have appropriate experience, in thiscase with the intended audience:199D 1 Why are we leaving [Accountexec 11 Out Of allthis stuff? (leans back and looks at M 1)M 1 (pause) Well--good question. He begged off ona lot of stuff because of his involvement inthe other project. Before he left Friday, Iwalked through content and Monday, I got thisback.D 1 Okay. So. We’ve got a lack of focus here.M 1 Well, umm--yeah.A 1 (nods vigorously) Let’s not hold back--we havea lack of focus.M 1 But I wasn’t upset about it after the resultsof this (points to section), because I figuredwith the time we’ve got now I don’t want towait for something-D 1 The kind of thing I’m thinking is--[Accountexec1] could review this. He’s known the clientfor over a year and knows what the currentenvironment is. He could say, “Oh no, thatstatement’s not true.” We could pick up whatthe right statement is.As strategist and chief drafter, Director 1 might perhapshave played more of a role not only in the early stages of theproposal development, but also in the middle stages. At thislate time, Director 1 now begins to re-cast much of theproposal, even re-visiting issues previously discussed inLevel I sessions among Manager 1, Technicians 1 and 2. Such arepetition of activities is obviously one reason that theproposal is so far behind schedule. One such activity is theclose reading of the RFP:D 1 How much of this structure is demanded by theRFP? Some of it? All of it?M 1 We can look in the RFP-D 1 This (points) is an unusual section.200M 1 (surprised) Is it? It’s a statement of ourunderstanding- -D 1 Yeah, I agree .... but to have it as a separatesection in a very prominent place is unusual.But I quite like the idea .... Why don’t we sitdown, scratch on the white board, get down ourpoints there. I’m probably not qualified to goaway and describe the points but maybe we canraise some questions and refer some questionsto some people and come up with answers fairlyquickly.Here we see Director 1 stepping into the role formerlyoccupied by Manager 1, reminding him of the importance ofcarefully reading and following the directions set out in theRFP in much the same way that Technicians 1 and 2 werereminded. Director 1 also re-asserts his authority as a LevelI drafter and a Level III strategist, while denying that hehas any Level II authority in technical knowledge. Later inthe same event, Director 1 tries to build a relationship withManager 1 as a more understanding collaborator. Teamrelationships are re-established here with Director 1admitting that his rules of proposal text production, forexample “Follow the RFP,” cannot be slavishly followed:M 1 I don’t think we’re far off with that structure[for the proposal].D 1 No. (laughs) They [Borden] don’t ask forpricing anywhere? (both laugh) This is fairlytypical. And they often don’t ask for asolution.What clients ask for and what they need are, apparently, oftenat odds. Director l’s self-deprecatory humour helps toestablish his role as a team member with the same difficulties201as other team members in focusing on the audience. He alsomakes valuable and expert comments about the shaping of theproposal draft which Manager 1 seems to need.Unfortunately, Director l’s comments concerning thestrangeness of client audiences in general trigger Manager 1to recall his problems with Jake, Borden’s proposal manager.A failure to deal effectively with a difficult audience hasalready slowed the production of the proposal, and heredistracts Manager 1 once again:M 1 I mean -- it [the RFP] is all mixed up! Undersection one, “Company Information”, he wants anoverview of the project and the projected timetable --D 1 And section 2 is executive summary? (renumbering and grouping, referring to RFPheadings) Okay.Director 1, who has not been a part of the draftingcollaboration until now, has also not shared in thefrustration which the team has often experienced in trying torespond to Jake. He does not reply to Manager l’s irritationhere over yet another example of Jake’s peculiar desires.Rather, he continues to follow the general drafting advice:“Follow the RFP” and do what the customer wants, as far as ispossible. When interviewed during this event, in response tomy question “Who is the audience for this proposal?” Director1 said “Mr. B.” Manager 1 said “Jake,” and then added “AndMr. B. And Director 1!” Audience focusing, it seems,continues to be a problem. However, once Director 1 takes202over the drafting leadership, his vision of the audience, ofcourse, prevails:D 1 Is there -- a lot of “sell words” --[Cerebellum] hype kind of stuff? Or are wemore into factual solution kind of stuff?M 1 What I meant -- what I thought you meant by“sell words” is -- are we explaining how thatsolution meets the customer’s need.D 1 Those are the right kind of sell words. Oftenthe sell words we put in there are hype “We’rethe best qualified company”. Very importantonce -- but after that it gets rather tiresome.M 1 I’m not sure it’s in there once!D 1 Once is in the letter on the front. We do moreof that with customers that don’t know us.Director l’s suggestions on helpful ways to focus on theaudience would certainly have been more useful much earlier inthe process. He no doubt assumed that Accountexec 1 wouldhave been more in the process than he was. An experiencedemployee could have kept Manager 1 better informed about thegeneral approach to the proposal which Cerebellum prefers.However, as a senior executive, Director 1 also bearsresponsibility in knowing the strengths and weaknesses ofemployees, and their other duties. For example, Accountexec 1was perhaps too committed with other work to give thesignificant assistance which Manager 1 needed to handle hisfirst Cerebellum proposal.Within an hour, Director 1 re-appears in Manager l’soffice with a general evaluation of his drafting. Not onlyhas Manager 1 written sections which now have been rejected by203Director 1, he has also adopted, according to Director 1, aninappropriate drafting style. Lack of leadership earlier inthe text production process is beginning to cost dearly:D 1 Too much justifying and explaining. We don’tneed a lot of words.14 1 Give me an example.D 1 “Minimize risk”. What does the followingsentence add? Maybe what kind of risk but --Right now it’s logic/persuasive driven. Itneeds sell. (enter Director 2 and Accountexec2) My temptation is to take it away to acorner, but we’ve got three people working onit -- we’ll get three different drafts. Let’snot panic and stay until midnight!D 2 [Because] then tomorrow will be a total writeoff.D 1 We’ll each take a chunk and spend an hourbefore bed with it. I’ll take four and one.(They divide sections. Director 1 and Director2 leave.)AE 2 (to M 1) It will come together.14 1 I should have redone this table of contentslast week to check this. I’ve done a lot Ican’t use.Valuable collaborations possible with sufficient timemust now be hurried in an attempt to maintain the schedule.More general expertise in drafting by all team members wouldcertainly help the text production process as well. Forexample, it would not have been assumed by the team thatDirector 1 would do all the “word-smithing” required, thusrelieving everyone else of the responsibility.In contrast, the Sundial proposal group are more able todiscuss their problems and come to a common understanding of204the difficulty and what needs to be done to address it. Suchbehaviour more clearly reflects a collaborative spirit inwriting the proposal. For example, Manager 2 realizes thatshe cannot successfully edit the document; however,Accountexec 3 takes the lead in assuring her that other teammembers will handle these presentational challenges as part ofthe proposal process. Such assistance from the accountexecutive marks one of the difficulties not resolved by theBorden team until a decisive final effort by Director 1 mustbe made to satisfactorily answer proposal presentationproblems. Here the collaborators both engage in and discussthe importance of text checking as part of Level I drafting:M 2 I don’t even know how that got in there; itdoesn’t even make proper English. All this hasto be gone through.T 4 There’s quite a few awkward sentences.M 2 It needs to be wordsmithed. That’s whatthey’re doing over there. That’s their job,that’s what they get paid to do. (waves to wordprocessing)T 4 I don’t know if they’ll check for grammar thatway.AE 3 Give it to [the General Manager’s assistant].Give her two days.M 2 Yeah, well, that’s what she said. I’m notchecking it for grammar or anything else. Isaid, “I’m French Canadian. They’re lucky Ispeak the language.”AE 3 By gar.M 2 So don’t count on me for propergrammartization(?) . ... Okay, so then we get tojust high-level business (continues to pointout sections to AE 3)205AE 3 (stops and takes a sheet. M 2 leans to see whathe is changing.) No longer passive! You couldget used to this, couldn’t you?1’l 2 (laughs) I’m assuming you’re going to go overevery last sentence.AE 3 Yes, and then [Director 5] will go over everylast sentence, and then [the General Manager’sassistant], and we may get [Director 4] to doit as well, considering the piles of money thatare going to roll inSuch a difference between the approaches of the Bordenteam and the Sundial group is more understandable given thatSundial reaches the final draft point some seven days beforethe deadline——Borden reaches the same point one day in advanceof submitting the proposal. Such collaborative activitiesrequire sufficient time to work properly.Event U. Level II. Rounding Up the StragglersAlthough Level I activities are now the most crucial tothe production of the proposal, collaborators still strugglewith specific problems resulting from a lack of information.As we have seen, Accountexec 1 has become a general reviewerof material, and Accountexec 2 now takes a very late butimportant role in information gathering and text composition.Here we see him working with Director 2 and Technician 1 tolocate and access missing prices from unco-operative vendors.Accountexec 2 has entered enthusiastically into his role asteam member in a number of ways. He not only has ideas aboutwhere to access information, but also is willing to actquickly to access a possible source rather than asking a206technician to do the job. Accountexec 2 is most familiar withthe employee in question and will likely be most effective atgaining his quick co-operation. However, since his authoritybase is not in technical knowledge, he must seek thecollaborative support of Director 2. The information hegleans from the branch office, however, is not immediatelyusable. Nevertheless, it does serve to remind Director 2 ofan indirect information gathering method. Technician 1quickly acknowledges and acts upon the new idea:AE 2 (to person on phone) We’re trying to get listpricing-- (looks at D 2 for assistance)D 2 (to AE 2) They [the vendor] won’t give us listpricing -- they won’t give us anything!AE 2 (listens to phone, then, to phone) Oh -- it’svery similar to [Soma]?D 2 (laughs) [Soma] what?!!AE 2 Yeah (laughs). (to phone) Technologically wedon’t know .... (continues)D 2 (suddenly excited, to T 1) Phone [Somal and askthem what their price would be with acomparableT 1 Okay! Give me the config!D 2 Hang on! Let me get the other piece of paper --we can fax it! (rushes of f)AE 2 (to phone) Do you have any contacts you mightbe able to tap? Any var’s [value addedretailers] you had relationships with?Such team work shows collaboration working at peak speedwith all team members entering enthusiastically intoactivities and doing jobs for which they are well suited.Accountexec 2 even continues to gather alternate sources for207information once Director 2 and Technician 1 leave to act on apossible lead, preparing, it seems, for the possibility thattheir efforts will be futile. He will also be able to use anyfurther information later on other proposals, and must becareful to treat his information source with respect.Event V. Level I. Words, Words, WordsDrafting continues well into the night on the day beforethe proposal is due. The technical experts (Director 2,Technicians 1 and 2) are busy talking through the solutionwhich will be presented to Borden. They work out the wordingwhile working out the technical solution. Technician 2 nowseems well entrenched as a respected member of the proposalteam. Here she and Director 2 succeed in sharing authority inan effective Level I collaboration:D 2 Usually we select -- uh, define -- testcriteria -- or acceptance criteria? To me --T 2 Don’t we have to first of all determine what itis we’re going to measure?D 2 So that’s what you mean by “test criteria”.Okay --T 2 What we’re going to measure and how we’re goingto measure. So, it should be broken into twothings then --D 2 Okay (writes).T 2 (looks at paper) What does that say?D 2 “Technical Commissioning”. I use the word“technical commissioning”, which means makingall the components work together (waves hishands and explains)T 2 Okay.208***D 2 Instead of “set-up”, can I say “configured”?T 2 Yeah, that’s no problemT 1 Configure original.D 2 (reading and writing) Configure networkapplication, set up system components. Yes.But, configure the application components --populate -- right.T 1 Yeah, that’s too vague. You do the proofingthere, right?Technician 2 has done much of the initial drafting of thetechnical solution and is prepared to justify her wording.Director 2 accepts her definition, and then adds to the draftand is in turn prepared to explain the reasons for his ownwording choice. Although Technician 2’s work on writing hasbeen at least partially due to Technician l’s reluctance todraft (leaving the job perhaps to a lower status member),changes during the late stages would appear to be of higherstatus. Both Directors 1 and 2 are involved in wordsmithingat this point, and Technician 2’s involvement in theconversation here shows that her opinions are respected.We see also the close relationship between drafting thetext and building the technical solution. Technician 1 isable to listen to Director 1 reading and adding to the textand make the evaluation that the text is vague. Such commentsindicate the emerging criteria for judging the textproduction. His query about doing the “proofing” seems toindicate that the wording may be vague because their209understanding is vague, and they need to talk further abouttheir plan for testing the hardware system.Event W. Level III. If OnlyLater still in the evening, Director 2 continues tostruggle with he wording of the proposal. Not only is hereviewing text but also making considerable additions to thetechnical sections of the proposal. Accountexec 2 hascontinued during this time to try to access the still missinginformation on vendor costs. He has managed to track down theelusive Accountexec 1 who has successfully contacted the onlyvendor still holding out on prices for hardware. Accountexec2 reports to Director 2 the results of the phone conversationwith the vendor:AE 2 Yeah, [George] said he didn’t like the way hewas treated, da da da da da da da da ... if wehad told them right up front that we were goingto do this, there would have been no bid fromtheir perspective.D 2 Shit. It would have been all over. They[Borden] woulda had one bid and we wouldn’thave had to worry or piss around with thisstupid thing (leans his head on the desk andlaughs).Although Director 2 can appreciate the irony of the situation,the proposal bid must still, unfortunately, be guided by thestrategy established during the early stages of the process.The general approach was re-formulated twice, but the finalplan put Cerebellum in what seemed to the other vendors to bean adversarial situation. If in fact, this last vendor had210known that their product would be tested, thus making them apossible team partner, then they would not have entered a bid,as they are now doing. It might have been a co-operativeeffort rather than a competitive one. The Cerebellumcollaborators have seen too late the problem with theirstrategy, and the results of failing to include, or evencommunicate clearly with, a possible team member. Director 2,Accountexec 1 and 2 decide that they have gone too far intothe bidding process now to try to change their approach.Event Y. Level I. Coming TogetherMuch later that evening, after many slices of pizza andmany hours of drafting and reviewing, the team tries to pullback together again and consider their product so far. Theysit together, going through the proposal page by page, tryingto see the proposal as a single coherent document instead of anumber of pieces, or a number of versions of one largedocument. Collaborators are here most clearly involved intext checking. A number of other activities are alsosuggested by their conversation. Role maintenance andschedule maintenance concerns are expressed by Director 1.Interestingly, although he is concerned with the best use oftime and the best use of personnel, he offers to be a wordprocessing operator (since the support staff has long sinceleft the office) to help deal with the problem of versioncontrol:211D 2 Okay. Next page. I like the companyinformation one. Very succinct,straightforward.D 1 Where’s [Accountexec 2]?D 2 He’s writing.M 1 He told me it [the letter] was the easiest oneto do. I said, “Is that like the guy who livesclosest to work: he’s always the last one toget there?”(group laughter)D 2 (to D 1) So, have you made any [changes todocument] --D 1 No. I honestly haven’t .... I wouldn’t mindplaying word processor [being a secretary] forawhile.(more editing discussion)D 1 We’re not making the best use of everyone’stime here.D 2 No, I know. Why don’t you go away and read?D 1 If you feel like that, I will. (smiles andleaves with one copy. Discussion continuesusing another copy.)This event in fact featured five people all trying to read thesame copy of the proposal document in aneffort to get allcomments on one master copy. Director 1 finally decides thatthe attempt to maintain one version of the proposal is notworth the loss of time such an activity seems to require, andretires to do more checking and .drafting on his own. The restof the group (Director 2, Manager 1, and Technicians 1 and 2)continue to check the text, and discover that much remains tobe done.212The activity of actually writing is particularly markedin this event. As I have earlier noted, remarkably little timeis spent in drafting, with most of the writing processdominated by various other events: such as talking, reading,searching for clip-text.Day ElevenOn this day the proposal is finally delivered. Last daydrafting includes a complete overhaul of the project workplan, review and re-drafting of the pricing section initiallydrafted by Director 1 during the very early hours of themorning, building and insertion of various colour diagrams,and of course, much editing, word-processing, photocopying,and assembling of proposal binders.Event Z. Levels III, II, I. End GameIt is difficult to construe the final day of proposalwriting as anything but one large event which moved throughlevels of engagement so quickly that collaborations oftenoccurred over a. spoken sentence or two, with a decision made,and then on to the next crisis. The air was tinged with panicbut also excitement as the group suddenly grew to include afull complement of support staff. A junior word processingoperator dissolved in tears and was replaced by a more senioroperator, the office manager shook her head with disapprovalover the number of last minute changes, the copy machineoperator misplaced the identification labels which belonged on213the proposal binders. Senior executives leaned on piles ofempty supply boxes (with the copier roaring at full tilt) andheld frantic conversations about pricing changes. The generalmanager’s administrative assistant insisted that even thereceptionist help by typing names on envelopes. Here then area few snapshots of that last day:M 1 Have you seen these prices?D 1 Well, I wrote it [the pricing section] . But Idon’t know as anyone else has seen it.The text checking process has almost broken down. However, aLevel III pricing meeting is clearly in order. Therefore,Accountexec 2 and Director 2 take the pricing page and checkit, coming up with a new pricing strategy which is thendiscussed with Director 1 in a hurried Level III meeting,conducted while standing in the hail while the first half ofthe proposal is already being reproduced.The senior executives retire to an office to read throughthe proposal in its latest incarnation:AE 2 You can’t have a five word opening statement.[“Borden is in a high growth market.”]D 1 (laughs) Oh, really.AE 2 A high growth market -- of what? Raspberries?At such a late hour, and in such a tense atmosphere,Accountexec 2’s attempt at introducing a presentation rule islikely not appreciated by Director 1 who, after all, draftedand re-drafted much of the text. It is clearly difficult forthe same collaborators to both build and check the same text.Accountexec 2’s role here as checker is a problematic one.214Collaborators hope, this close to the deadline, that his jobwill be limited to finding typographical mistakes andmislabelled diagrams. The tension here may also be caused bythe group of senior employees who usually find themselvestogether in Level III strategy meetings now thrust into aLevel I activity. The levels seem to collapse as, finally, itall comes together. Authority now is based on the person withthe best presentational knowledge, who also happens to be theexecutive with the highest status: Director 1. Members seemgenerally on edge, but disagreement is at a minimum.Most amazing of all is the final example of Level IIactivity. The copier is running for the final time,punctuated by cries from the senior word processor of “Stopthe machine!” every time she is asked to make one more changeand a page must be re-copied. The final version,approximately the fifth (though some sections have been reworked more than others) totals twenty-eight pages. Bindersalso contain the resumes of the technical staff (Director 2,Technicians 1 and 2), the Cerebellum annual report, and acovering letter (one copy each to Jake and Mr. B).Suddenly, the fax machine beeps to life, and the numbersarrive from the last vendor giving list price on theirhardware configuration. Even more astounding, these numbersexactly match those numbers which Director 2 estimated andwrote into the proposal mere hours before. Technicians 1 and2 check the figures, share them with Director 2 and generallyenjoy a moment of success.215Twenty-five minutes before the proposal is due,Technician 1, laden with proposal binders, leaves for Borden’soffice which is located forty-five minutes away. He deliversthe proposal twenty-three minutes later, and phones Cerebellumto report success. Accountexec 1 is relieved. The rest ofthe team has already dispersed—gone home or moved on to otheractivities.Finally, then, we have witnessed in these three chaptersa detailed chronological account of the collaborative textproduction efforts surrounding the writing of a businessproposal. Early events were dominated by the struggle todevelop a winning strategy, what I have called Level IIIactivities. The strategy sessions were fed by informationgathering activities, what I have called Level II. During themiddle events, drafting sessions come to the fore, thoughthese are still mediated by Level II information activities.Final events show a continuing emphasis on Level I drafting,with some Level II activities continuing, and a return toLevel III strategy sessions. These later Level III eventsfocus on approaches to the rhetoric of the proposal. Thelevels finally dissolve in the last burst of energy tocomplete the proposal packages.A PostscriptThe Borden proposal was initially rejected. The generalstrategy of offering a number of vendors as optional partnersproved to be unsettling for the client. The uncertainty of216not knowing exactly who they were dealing with caused them toaccept the bid from Sauct. However, after twelve months ofdealing with the other company, Borden called Cerebellum andasked them for assistance. Cerebellum was able to enterBorden as systems integrators, co-ordinating the work of Sauctand adding other system components (such as project managers).In the interim, Accountexec 1 and Director 1 had leftCerebellum. Nevertheless, Accountexec 2 and Technician 1initiated plans to form a Cerebellum project team to helpBorden with its business needs.Sundial did not accept Cerebellum’s proposal. Sundialaccepted a higher priced bid from a firm which had extensive“imaging” experience in British Columbia, which Cerebellumlacked. Cerebellum continues to enjoy an excellent long-termrelationship with Sundial and hopes to have future businessopportunities with them.217CHAPTER EIGHT: CONCLUSIONS[A]nother beginning, a chance to continue theconversation pursued in and between these pages withyou. (Lisa Ede and Andrea Lunsford, SingularText/Plural Authors, p. 143)The collaborative text production process istremendously complex, and, at first, seems to refuse allattempts at discovering pattern and lesson. Myethnographically-based research with the writers of Cerebellumreveals the intricate social dynamics of writers at all levelsof the company hierarchy as they interact to write a businessproposal. Cerebellum writers construct the text together,both face-to-face and in concert with others who may be absentin time and place and represented in previously written texts.Various activities enhance or impede the collaborative writingprocess at every level of engagement and throughout everyphase of the text production process. A close analysis ofthese activities helps us to see more clearly the socialconstruction of the document.The focus has remained steadily on text, as it isnegotiated by collaborators. That writing is indeed a socialprocess is perhaps nothing new. We see in this study,however, that sophisticated drafting skills are necessary inthe group writing project. As much as writers constantlyrehearse and discuss ideas, they seem far stronger atdeveloping strategy and gathering information than in actualtext drafting. However, our understanding of the nature of218that process obviously needs continuing in-depth study, and away to deal with the resulting details still perplexing intheir intricacy. The way I have chosen to show the proposalwriting process at Cerebellum, in a form more manageable andaccessible by business writers, is by constructing a model,which, though in need of further testing, begins the job ofrepresenting the collaborative writing process as levels ofengagement with text. Thus, I hope to make furtherconversation on the “multivocal, multiplicitous collaborativewriter/text” (Ede and Lunsford, 1990, p. 143) more open to myinformants.The preceding three chapters have described and discussedsome twenty-three events. These collaborative activities areundertaken by a core group of eight employees writing abusiness proposal. My focus has been the development of theBorden proposal, supplemented and corroborated with eventsfrom the Sundial proposal effort. Such a re-telling shouldgive readers a sense of the narrative of the proposal writing,a drama complete with heroes and villains, conflicts andclimaxes, successes and failures. However, in an attempt torepresent the collaboration in a manner more helpful to thosewho would seek to analyze and then intervene productively inimproving collaborative text production, I also offer acategory system of levels. The Levels of Engagement withText, developed after extensive viewing and reviewing ofvideo-taped events, will serve, I hope, to begin the process219of ordering some of the complexity inherent in the socialprocess of writing.Further to the end of informing my various possibleaudiences, in both academe and business, I will now drawtogether some conclusions by looking at patterns discerniblein the events assigned to each level. All writing is writing,and yet has specific demands, whether a proposal, schoolreport or a poem. Here, I address my comments as possiblelessons for teachers of writing in various settings. Onceagain, as in my introduction to the levels in Chapter Four, Iwill discuss Level III, then II, then I. The progressionrepresents the appearance of the level in the chronology (thatis, proposal efforts begin at Level III), and is the sequencemost familiar to the collaborative writers under examination.I have maintained the same presentation order in each level sothat readers might more easily compare comments under varioustopics: level objective, issues of time, lists of buildingand checking activities, personnel, and finally behaviourswhich enhanced and impeded collaboration.Level III FindingsEvents are classified as being at Level III if the mainobjective of the meeting is to discuss the general strategywhich will be undertaken to win the bid. Examples are thecollaborators’ decision during Event A (The Starting Gun) thatCerebellum partner with Sauct, and Event G (Quick ChangeArtistry) in which they establish three optional partners,220clearly favouring an association with Azure. General businessstrategies, it seems, may be simply stated in a singlesentence, much like the controlling thesis in an essay (andmuch like theses, are rehearsed and repeated as well)Cerebellum employees call such statements “high-level”declarations to emphasize their function as general headings.Strategy decisions apply not only to questions of the vendorpartnership described in the proposal, but also to the generalrhetorical flavour in which the partnership is presented.Event T (Asking the Tough Questions) shows how a rhetoricalstrategy structured in logical arguments gives way to a sales-driven approach.Major changes in business strategy cause problemsthroughout the proposal. Collaborative Level III events(strategy discussions) dominate the early days of the proposalconstruction, completely disappear in the middle phase andthen re-appear late in the text production process. Commonsense might indicate that a strategy should be decided uponfirst, and then the rest of the process might be taken up withdetailing the strategy within the proposal text. However,during the Borden proposal, there are two shifts in thegeneral direction of the business strategy (for a total ofthree) . These shifts dominate early events. During lateevents, for example, frustration is revealed in theconversation between Director 2 and Accountexec 2 in Event W(If Only) when they bemoan their Level III decision, and inlevel jumping by Technician 1 during Event D (Who’s On First?)221when he ignores a Level II meeting objective to criticize thebusiness approach in a Level III manner. Finally, I note thataccording to Director 2, the business strategy was acontributing factor which caused Borden to reject theCerebellum bid.A late shift in the rhetorical strategy, which improvesthe text’s selling focus, also causes a loss of time andincrease in anxiety. During Event T (Asking the ToughQuestions), Director 1 rejects Manager l’s rhetoricalapproach, which focuses on lengthy arguments for the proposedsolution, and also assigns re-writing roles to ensure that theproposal will be completed in timely and efficient fashion.This event marks a change in rhetorical strategy (for a totalof two different approaches). Such recursion would seem to bea feature of collaborative writing, unlike the more linearprocesses of some solitary writers such as the engineerobserved by Selzer (1983) . Returning to jobs thought completeis perhaps a function of the large number of people who mustbe satisfied with the strategy. As some informants in othercollaboration research have commented, collaboration can oftentake more time than would a process undertaken by a singlewriter, however, the superior quality of the final documentmakes group work worthwhile (see also Allen et al., 1987;Bryan, 1992)A lack of collaborative Level III events during themiddle of the process is perhaps a result of seniormanagement, especially Director 1, assuming that Accountexec 1222had taken leadership or at least a major consulting role inoverseeing the proposal. Such ownership is no doubt inferredby Director 1 since the role of the account executive duringany proposal is to assume final responsibility. In the caseof the Borden proposal, such responsibility is assumed byDirector 1 especially because of promises made by Accountexec1 during Event A (The Starting Gun) that Manager 1, new toCerebellum, will receive his significant assistance.Accountexec l’s reluctance to enter fully into thecollaborative process creates tension within the Bordenproposal group. Such an outcome might be inferred from thework of organizational communication researchers (such asStohi and Schell, 1991)In summary, then, I have discussed five Level III eventsduring which the general strategy is discussed, with primaryemphasis placed on idea generation and trying to differentiatethe Cerebellum approach from those of possible competitors.Ideas are collaboratively evaluated, modified and elaborated,in order to reach agreement on a single, high-level approach.Level III meetings may be further classified into “building”and “checking” events.Business strategies seem to be adopted for a number ofreasons. In the Borden and Sundial proposals, we see examplesof approaches based on the most powerful technology, the leastrisky solution, and the most politically astute strategy. Thefirst shift in strategic approach during the Borden proposalis marked by a very informal meeting in the hail, which I have223called Event C (Changing Horses). Apparently, a new strategyis being “checked” (evaluated and found attractive enough towarrant a change in direction). The original strategy, basedon the strongest technical solution has been rejected infavour of evaluating three possible partners which will reducethe risk of a possible misfit with Sauct. The second strategyundergoes final modification and elaboration during Event G(Quick Change Artistry).The final approach retains the testing of optionalpartners but adds the politically motivated preferentialtreatment given to Azure, which will be tested first andadopted if satisfactory with no further testing of otherpotential partners. Event W (If Only), most obviously a“checking” event, indicates the further evaluation of thestrategy enshrined during Event G (Quick Change Artistry).The strategy is now found by Director 2 and Accountexec 2 tobe problematic at best, but the closeness of the deadline datemakes further changes inadvisable. Such behaviour may beattributed to the general tendency notedin individuals andgroups to persist in “losing causes” once a great deal of timehas been invested (Staw and Ross, 1989).Event T (Asking the Tough Questions), the only Level IIIevent discovered which deals with rhetorical approach, mayalso be called a “checking” event (of one strategy) whichlater becomes a “building” event (of another strategy) . Ofcrucial interest here is Director l’s approach: he evaluatesand then modifies the style of the proposal to reflect more224“sell” (for example, fewer words to increase speed and ease ofreading) . Finally, Event Z (Endgame), notable for itscollapsing of levels, features a short and hurried checking ofpricing, which will be one of Cerebellum’s major sellingstrategies, followed by evaluation and then modification ofthe price quoted.Certain patterns can be noticed as well in the personnelwho attend Level III events. All Level III meetings demandthe attendance of high status employees, but the largenumbers of executives involved in the early phase drop topairs making high stakes decisions in the late phase.Director 2 appears at four of the five events, as does Manager1. Other executives (Dl, AE1, AE2) appear three times.Director 3 appears twice and the General Manager once.If quantity alone were considered, it is notable thatAccountexec 1, who bears ultimate responsibility, is notpresent as often as Director 2 and Manager 1. Director 2appears most often, quite surprising for such a senioremployee with responsibilities on several projects. Totalnumbers for early and late Level III events also markedlycontrast. Early events feature large numbers of employees atmeetings: Event A (The Starting Gun), five people; Event C(Changing Horses), four; and Event G (Quick Change Artistry),seven. Late events (T and W) at Level III have only pairs atwork together.225If meeting topics and significance are considered, otherpossible conclusions might be reached. One might speculatethat early strategy decisions are important, important enoughto involve all of the most experienced employees in a verypublic performance of power and knowledge. Early strategymeetings also seem to carry high status—being left out,perhaps, is a diminution of power. Late strategy meetingssuch as Events T (Asking the Tough Questions) and W (If Only)are hurried and best accomplished in small groups, where twocan agree and quickly move on. Events T and W involveDirectors 1 and 2, Manager 1 and Accountexec 2. They form thenucleus of Level III personnel during the final days of textproduction.Collaborative activities during Level III events showbehaviours which enhance text production and those whichimpede it. Topics which I will address are teamrelationships, audience sensitivity, schedule maintenance,process awareness, and level flexibility.Team relationships at Level III events function largelyto support text production. The initial meeting, Event A (TheStarting Gun), shows in some detail the dynamics of Level IIIcollaboration. It involves all three directors, the proposalmanager and the account executive (whom I have calledAccountexec 1). Although a hierarchy exists (for example,Director 1 has the final word), these executives work togetherin a loosely-structured meeting, generating ideas and optionsfor group consideration. Team discipline is lax with free-226wheeling conversations and brainstorming sessions occurringfrequently with the focus on holding the group together.Roles are kept largely undefined and everyone is encouraged toparticipate in at least a semblance of equality.Two recurring subthemes of building and maintaining teamrelationships are the search for differentiation and the useof humour. An important way to form a team for Cerebellumseems to be in establishing a group identity, in answering thequestion, “who are we?” One way to accomplish such a task isto ask, “who are they?” and then use the differences toidentify the “home” team. Several examples of this behaviouroccur during Event A (The Starting Gun). Competitors arediscussed and labelled. One group is found to be “sleazy” forfailing to return a business favour. The implication seems tobe that “we” as Cerebellum employees agree that such behaviouris reprehensible and are more united because we agree. Thisis not to say, however, that group decisions ondifferentiation always work to the benefit of the team. Forexample, the unique strategy developed by Level III meetingsin Event C (Changing Horses) and G (Quick Change Artistry)results in a system of optional partners which the clientfinds too risky, too “different” to be attractive. Therevealing discussion in Event W (If Only), where Director 2and Accountexec 2 lament the final strategy, marks a teamdetermined to work together until the end, united at least ina losing cause.227The sub-agenda of a power struggle, on the playing fieldof text production, runs throughout Level III meetings,sometimes assisting the maintenance of team relationships andat other times threatening it. Obvious uses of authoritybased on company status are notable in Event A (The StartingGun) and Event.T (Asking the Tough Questions). In Event A(The Starting Gun), Director 3 ‘s verbal and nonverbalbehaviour threatens the role of Accountexec 1 as the proposalleader. His experience in sales matters is lost when theaccount executive fails to take note of his ideas, for examplein dealing with Jake’s aversion to risk. In contrast,Director l’s forceful handling of Accountexec l’s lack of“focus,” discussed in Event T (Asking the Tough Questions),serves to re-direct the team to meet their deadline when heassigns Accountexec 1 a new role as reviewer and allowsDirector 2 to continue in a general leadership capacity.Cross (1990) also notes that the presence of a group hierarchyis a convertible factor--it may tend to draw a group together,or tend to force it apart.Humour also assists the text production process duringLevel III. The ironic laughter shared by Director 2 andAccountexec 2 during Event W (If Only) serves to relieve thetension produced by a text well behind schedule, and unitesthem for the moment in sharing a joke. Similar behaviour canbe found in Event A (The Starting Gun), the Sundial proposal,when the bantering surrounding “he can do it both ways” allowsa group excited about a Cerebellum success (albeit on another228contract) to share laughter. Humour also seems to regulatebehaviour within a group, to warn team members aboutinappropriate actions. When Director 3 suddenly involves anunprepared technician in the Sundial discussion during Event A(The Starting Gun), the jokes and laughter remind him that hehas erred, and also ease the discomfort of the haplesstechnical expert newly drafted onto the proposal team.Building a shared notion of audience, which will be borneout by the proposal text, has an impact upon teamrelationships as well. Since, however, Cerebellum’s attitudeteeters between an adversarial and an inclusive role for theaudience, I will consider it separately. Three examples showLevel III collaborators trying to deal with the audience. Inthe Borden effort, Event W (If Only) marks a final realizationat Level III that the vendors have not been treated well byCerebellum (a realization which Technician 1 voiced in LevelII and I meetings very early in the process) . Ironically, a“red-flag” was waved as early as Event C (Changing Horses)when Director 2 says: “This situation isn’t fair to ourpossible partners,” but the vendors, as another audience whichcould have been included on the team, have instead beenshunted aside by competitive concerns.The treatment of the client company as audience proves tobe a challenge throughout the Borden proposal writing process.Cerebellum tries two quite distinct approaches which they useto imagine their readers: audience as specific person, andaudience as general abstraction. Articulation problems in the229early and middle phase seem to be centred on history of theCerebellum in dealing with a specific person: Borden’sproject manager, Jake. During the current proposal, Manager1, influenced by Technicians 1 and 2, gradually conceives ofJake as an adversary. By the time he discusses audience withDirector 1 during Event T (Asking the Tough Questions), verylate in the process, his frustration with Jake has almostoverwhelmed him. Director 1 can offer a general view of theaudience and even joke about the difficulties of all clients.From this point forward, a more dispassionate view of audiencedominates the text production effort. In contrast, theSundial proposal offers a clear and sympathetic view of aspecific audience (“Steve”) from the first Level III meeting,as constructed for the team by Accountexec 3. Accountexec 3reports that Steve “imagines the project as a partnershipbetween [Sundial] and the vendor [hopefully Cerebellum)This is an actual, identifiable person interested in being amember of a team, and invited to be one by a Cerebellumemployee. Inviting the audience to join the community is arhetorical strategy advised by scholars such as Perelman(1982)Clearly not a characteristic activity, the only recordedinstance of schedule maintenance at Level III occurs duringEvent T (Asking the Tough Questions). Such infrequency may bea result of the general quality of the Level III meeting: itis free-ranging and relatively undisciplined, encouragingactive participation and idea generation. The exception230occurs during a markedly hierarchical meeting wherein Director1 inquires about Accountexec l’s involvement (“Time might beall”). Manager 1 misunderstands, thinking perhaps thatDirector 1 wishes to reintroduce Accountexec 1 as leader (“Ifigured with the time we’ve got now I don’t want to wait forsomething”). In the interests of schedule maintenancehowever, Director 1 is proposing that Accountexec 1 become atext reviewer. Time pressures help the team to stay on task,but also work against the thorough examination of the bestpossibilities for text development.Overall process awareness is shown in Level III Events A(The Starting Gun) and T (Asking the Tough Questions).Participants in Event A point to level I drafting. Director 2instructs Accountexec 1 to provide examples of proposals, andAccountexec 1 instead gives names for Manager 1 to get forhimself. These samples provide models of “structure” and“technical content.” Event T (Asking the Tough Questions)also shows strong awareness of Level I concerns, which is notsurprising given that the proposal is almost due. Event A(The Starting Gun) also points forward to Level II informationgathering concerns. Director 3 asks Accountexec 1 to seekinformation from another vendor involved with Borden (“Seewhat you can get”)Perhaps because Level III tasks are generally highstatus, and other levels are of lower status, levelmaintenance concerns, that is, comments designed to ensurethat the team keep to the current meeting objective, are231almost absent during strategy collaborations. As previouslymentioned, discipline is generally lax at Level III, withparticipants usually not engaged in imposing restrictions oneach other. Scant evidence was found of Level IIIcollaborators cautioning team members to remain on task.High status employees would almost never be tempted, it seems,to jump to Level II, information gathering, or Level I,drafting. The only example of such behaviour comes duringthe Sundial proposal, Event A (The Starting Gun), whenDirector 3 contacts the outside technician by phone, a jump tolevel II. However, the account executive reprimands thedirector for his action indirectly with humour (as discussedabove)Level II FindingsEvents were classified as being at Level II if the mainobjective of the collaboration was to obtain or discussinformation. Examples are Accountexec 1 drawing up a proposaltimeline to inform Manager 1 during Event B (Opening Moves),and the display of estimating by Director 2, Event S (PullingRabbits out of Hats), which is based on a consideration ofinformation gathered thus far and offered for checking byothers. Level II reaches out beyond the immediate teamfurther into the company, and then beyond into othercompanies, government documents, or wherever material might befound. Information, it seems, supports and mediates allproposal writing activities, and is, indeed, treated as an232important commodity. It is usually traded for future businessopportunities or for more information. When access toinformation is blocked, it can seriously hamper a company’sability to build a proposal, as we see during the Borden bideffort (see especially Event U Rounding Up the Stragglers).Collaborative Level II events play a strong mediatingrole throughout the proposal process during early, middle andlate phases. Level II events are more frequent, however,during the early phase, as might be expected. Most of theinformation needed to decide on a strategy is discoveredduring Days Two and Three. Other information gathering occursin tandem with the building of the solution for the client’sbusiness problem. Information crises continue to occur evenvery late in the text production process: witness especiallythe costing figures arriving and being discussed during EventZ (Endgame) in the final hours before the deadline.Throughout the entire time spent on the proposal, Level IIactivities continue to serve as intermediary events, informingstrategy decisions and drafting approaches. Readers aredirected especially to consider the rhythm of movement betweenLevels, for example, the shunting between III and II duringthe early days, and that between II and I during middle andlate phases. Event F (Under Consideration) especially shows ameeting which, although intended to be at Level II, begins atLevel III and ends at Level I.233To reiterate, then, I have discussed eight Level IIcollaborations in which collaborators are gathering orconsidering pieces of information which may later be addedinto the text. Readers will note also that there are moreLevel II events than Level III. Even more Level II eventscould have been observed and discussed had not companysecurity issues intervened. Technicians were reluctant tohave me record conversations with vendors. Also, many ofthese events occurred off-site in vendor offices. Strategydiscussions are less frequent than are conversations toacquire and consider information, even in the Borden proposaleffort where the strategy undergoes several changes.These eight Level II events may be considered as eitherbuilding or checking conversations. However, such activities,as we saw above in Level III, are not entirely discrete. Twoclear examples of information building may be seen in Events Band N. •Event B (Opening Moves), already discussed above,concerns Manager 1 asking for and receiving schedulinginformation from Accountexec 1. The gathering of suchinformation would probably not be necessary except thatManager 1 is new to Cerebellum and its business approaches.Since Manager 2 finds herself in the same position as anewcomer during the Sundial proposal, however, I speculatethat such information seeking may be fairly common if acompany prefers to hire necessary personnel only when requiredas Cerebellum does.234During Event B (Opening Moves), information is alsosought outside the branch office by Accountexec 1. Thesource, Technician 6, proves reluctant to provide informationbut eventually does so. Event N (Collabo-tech) is anexcellent example of a senior technician (Technician 1)seeking and receiving information from an in-house expert(Technician 3) who is more junior in the company hierarchy,but more knowledgeable on a particular technical issue. Inboth events, conversants engage in more than a simple questionand answer session, discussing problems (Event N) andinterpreting documents (Event B) collaboratively.Information is reported and often comes under furtherconsideration, briefly or at length but always with majorimpact on the text, during what I have termed “checkingmeetings” at Events E, F and L. Event E (Decisive Action)concerns a brief report by Manager 1 to Accountexec 2 aboutinformation on a software product which is not available,contrary to what the team previously thought. Thisinformation results in a meeting with Mr. B. to keep himinformed and changes the proposal’s technical solution.Manager 1 reports information of great concern on Jake,Borden’s project manager, to Technician 1 and Accountexec 1during Event L (The Audience Awaits). He seems to be checkinghis negative reaction to Jake with them, perhaps forconfirmation, but also trying to build a picture of what Mr.B. might value in a proposal.235Event F (Under Consideration) is the longest Level IImeeting on record at 178 minutes and another event primarilygiven over to checking. Here Technicians 1 and 2 arereporting on their work thus far in building the technicalsolution. Director 2 is reviewing their information incollaboration with others at the meeting.Because they occur so late in the process, Events S and Uare subject to time pressures which may be beginning to causea convergence of the functions of building and checking. Inmuch the same way, the levels themselves begin to collapseonly a short time later. Cross (1990) considers time to be afactor which draws groups together. However, my researchshows that a lack of time can also work against the buildingof team roles. For example, additional time might haveprovided opportunities for junior members such as Technician 2to learn more about working together on the information-gathering processes. Event S (Pulling Rabbits out of Hats),for example, involves Director 2, Manager 1 and Technicians 1and 2 developing and considering estimates of time needed tocomplete the Borden project. Event U (Rounding Up theStragglers) considers a small piece of information arisingfrom a telephone call initiated by Accountexec 2, and showsIDirector 2 and Technician 1 using the resulting idea to seekvendor information from an alternate source. They arebuilding, checking and then building again.236Consideration of Level II events must also include therather aberrant Event D (Who’s On First), in which Technician1 accosts Manager 1 looking for information. However, we mustcall this a “false” Level II, because Technician l’s realagenda seems to be the resolution of an argument over strategysuitable only for a Level III meeting. The conversation’spurpose, as established by Technician 1, is to air agrievance, or perhaps affix blame, and not to seek informationon proposal writing issues. Event Z (Endgame) also containselements of Level II during the final remarkable arrival ofthe costing information. The last vendor has relented,perhaps hoping that the information will not be more thanminimally useful but unwilling to make a permanent enemy ofCerebellum.An examination of the personnel present at Level IIevents shows a division of labour. Many of the informationgathering phone calls and meetings were conducted by thetechnical staff, especially Technician 1. My data shows himat six of the eight events I observed, and he also conductedmuch of the work on Days Two and Three. Technician 1 worksquickly and effectively. He seems to know his role on a teamand such knowledge may help him to work effectively as acollaborator (see also Newman, 1988).Manager 1 also appears at six events in a generalcapacity as proposal manager. Director 2 appears at threeevents as both technical expert and leader, as doesAccountexec 1 as client expert. Accountexec 2, serving as237substitute writer and sales expert, and Technician 2, workingas technical assistant, each appear twice. Two othertechnicians, 3 and 6, also provide information on one occasioneach. Two team members, Directors 1 and 3, never appear atLevel II meetings, perhaps feeling that they have no expertiseat either gathering or considering information, especially asmost of the ideas under discussion at Level II are fairlytechnical. Even Manager 1 has difficulties coping with thislevel of information (see, for example, Event P Enter theDragons)Directors seem to feel that their role is to launch theproposal, and they only step in during the late phase when itseems in danger of floundering. A further look at kind andplacement of meetings attended shows that Director 2 appearsonly at checking meetings (Events F, S, U) and he does notappear during middle phase Level II meetings. (In fact, nodirector appears at any middle phase meetings at any level!)More attention by Directors during middle events may havesolved problems of information consideration from a moreexperienced point of view much earlier in the text productionprocess. Unlike Level III collaborations, Level II meetingsremain fairly consistent in size; that is, late phase meetingsare generally as large as early meetings. Information, itseems, retains its power to influence proceedings.As in Level III, collaborative activities at Level IIreveal behaviours which enhance text production and thosewhich impede it. Team relationships are both strengthened and238threatened, audience is considered wisely or unwisely,schedules are maintained or ignored. Collaborators show bothflexibility and inflexibility in persisting with Level IItasks.Team relationships at Level II are fairly pragmatic.Sources of information are sought and discussed quickly duringbuilding events. The goal is straightforward, “tell me whatI need to know.” Technicians 3 and 6 (Events N and B) arecompany employees invited to collaborate briefly and thenreturn to other work. Information sought from vendors who arenot controlled by Cerebellum is not quickly supplied, becauseof vendor annoyance over Cerebellum’s approach (see Events Land U). The cut and dried free market of information iscertainly tempered by more volatile notions of “fairplay” andtrust.Power and authority are held at Level II buildingmeetings by persons with information, irrespective of theirstatus (see Events N and U) . Strains in the hierarchy occur,however. Events B and D show attempts by less knowledgeablepeople to put themselves on an equal or superior plane withthe more knowledgeable employee. In each of these events, aquestion is asked which is less a request for information thana challenge. During Event B (Opening Moves), Manager 1 asksAccountexec 1: “What happens two years down the road when theyhave to spend another million?” and in Event D (Who’s OnFirst), Technician 1 quizzes Manager 1: “How will we maintainour credibility with vendors?”239Often, however, the hierarchy at Level II appears to bethe same as at Level III (by status) because the most senioremployees also often have the most knowledge and experience(see Events B, E and F, which show the experience ofAccountexecs 1 and 2, and the technical knowledge of Director2) . Readers should also note that the high status, experienceand knowledge of Director 2, in conjunction with his dominant,extroverted personality tends to put him in the leadershiprole at any Level II meeting he attends. When time is shortand the list of tasks long and complex, Director 2 does anundeniably remarkable job. His approach, however, does notlend itself to promoting the ideas of junior team members andhelping them to become stronger team members. Often, as Veiga(1991) has noted, the presence of such an employee can causeother team members to contribute less.The most stalwart collaborators, who usually work onfairly equal footing, are Technicians 1 and 2. Several stepsseparate them officially on the Cerebellum hierarchy: one is asenior engineer and the other is a junior associate engineer.As I have noted, Technician 2 finds it challenging to work herway onto the team especially at Level I checking meetings.However, Technicians 1 and 2 work together at Level II, bothalone and with others, without the hierarchical quality soobvious in most Level II collaborations. They share tasks,support each other during meetings, and even use the pluralpronoun “we” to report their findings or describe their workto date at Level II checking meetings. They work together in240a more dialogic (Ede and Lunsford, 1990) fashion. Pastexperiences too may contribute to their effectiveness as ateam (Tebeaux, 1990).Audience concerns, focusing on the difficulties ofimagining the enigmatic Mr. B or the unreliable Jake as theprimary reader, permeate Level II. Although Kirsch (1991)showed how experienced writers compose for audiences whichthey perceive as either above or below them in the samecommunity, clearly the formulation of an audience for thesecollaborators is a more problematic activity. They havetrouble reaching an agreement to decide if the audience is tobe represented as within their community, or even whether theaudience may be considered as above (as Mr. B. seems to be) orbelow (as Jake appears to be).Event B (Opening Moves) concerns Manager 1 accessingAccountexec 1 as an information source knowledgeable aboutBorden and Mr. B. Accountexec 1 seems to base his replies onprevious experience with Mr. B., rather than on theinformation contained in the RFP. For example, he wantsCerebellum to avoid taking any risks associated with theproject, but later in the same meeting informs Manager 1 thatBorden must be told in the proposal that Cerebellum’smanagement of the project will “minimize” their risk. No one,it seems, will be at risk, which Manager 1 seems to havedifficulty believing (“So in the end, they take the risk”.)241In contrast, when Manager 1 takes a problem toAccountexec 2, his handling of audience is more sensitive.Event E (Decisive Action) shows Manager 1 concerned thatsoftware needed for the project will not be available asBorden has been lead to believe. Accountexec 2 immediatelysuggests a meeting with Mr. B., which will include Jake, atMr. B.’s discretion. The owner will be kept informed of newdevelopments, and Cerebellum will avoid unpleasant surprises,or the possibility of alienating Borden’s quixotic projectmanager.Still, however, Manager 1 must use Jake as a source ofinformation on which he will base his writing of the proposaltext. Event L (The Audience Awaits) shows the frustrationManager 1 feels at having to use Jake’s facts to inform theproposal writing process. He strongly suspects that Jake isinventing the rules as he goes, even vital information such ashow the text is to be judged. Jake becomes the obstructingaudience which can be neither ignored nor dealt with as areliable source of information. He has been alternatelyscorned as a project manager and consulted as a representativeof Mr. B. Later in the same meeting, when information isfurther discussed, Manager 1 suggests that Jake and Mr. B.have different priorities, further complicating the proposaleffort. At this point, Manager 1 seems to be more concernedwith the reaction of the “senior executive,” Mr. B., whom, hehopes, has reactions typical of executives in general: “Theywant to make little decisions.” Presumably, consensus on such242information by the collaborators present in this meeting willmake targeting the audience easier. It will also help toshape the proposal into a series of sections which argue smallpoints one after the other.Although there is often a tone of urgency in voices, forexample in Event E (Decisive Action) (Manager 1: “How can weguarantee software if it’s not up yet?!”) and Event U(Rounding Up the Stragglers) (Technician 1: “Okay! Give me theconfig!”), schedule maintenance is only infrequently mentionedat Level II as a direct reminder from one team member toanother that they must hurry. Event N (Collabo-tech) showsTechnician 1 consulting Technician 3, a very peripheral teammember, and rejecting his first idea because it would take toomuch time: “We’ve got two weeks max.” A far more directreminder comes from Director 2 in Event F (UnderConsideration) when he pushes for completion of networkpricing: “Okay. So we’re going to have a network priced out,rough cut, either today or tomorrow,” even though such work isfar from complete. Technician 1, in face, quickly says:“Tomorrow,’1 and as we know from later events, networks werestill being priced out on the last day.Being aware of the process as a whole becomes a topic ofconversation in Events B, F and L. Manager 1, during Event B(Opening Moves), shows that although the prime objective ofthe meeting is to gather information, he is cognizant of bothLevel III and Level I concerns. Such awareness shows onceagain the mediating role played by Level II activities. For243example, he worries that “This is a high-risk project,!! astrategy concern, and also wants to discuss scheduling for thedrafting activities. Event F (Under Consideration) shows bothDirector 2 and Manager 1 reminding the team of Level I,drafting, activities; for example: “I can start outliningcontent with Director 1 at 10 o’clock” (Manager 1). Finally,Event L (The Audience Awaits) shows Manager 1 considering theimpact of information on future Level I activities, draftingthe proposal to assure maximum effectiveness. He leads acollaboration which agrees on a likely list of selectioncriteria for the proposal’s acceptance by Borden’s owner:costs and flexibility.Maintaining the level, that is, focusing on the task athand, is more obvious at Level II than Level III. Techniciansespecially are most comfortable at Level II and rarely seek tojump to another level (though they often want to return toLevel II when in Level I). One notable exception isTechnician l’s behaviour during Event D (Who’s On First),which, as we have seen already, sounds like an informationsession, but the information Technician 1 desires seemsintended more to attack the strategy (a Level III concern)than to advance the proposal.Such level-jumping has an impact upon the text productionprocess. Other collaborators are also given to shifts infocus as we see in Event F (Under Consideration) . HereDirector 2 and Accountexec 1 struggle to remain in Level III,bantering about strategy, rather than focusing on information244consideration which should be the goal of a Level IIcheckpoint meeting. It is interesting to note thatcollaborators seek to jump up to Level III and not down toLevel I.Some further consideration of level-jumping and levelmaintenance in the model would seem to be appropriate here.Flower and Hayes (1981) consider recursion to be a valuablepart of the cognitive writing process as it applies to theindividual writer. Recursion during a group meeting, however,is more difficult than would be the case with a solitarywriter. The collaborative group is a social organism. If onemember disrupts the meeting by trying to jump to anotherlevel, then the group’s progress as a group is damaged, as isprogress on the document. Each level of collaborativeengagement is important but different. Level III meetings,for example, need Level III members such as directors to meettheir strategy objectives. Level II meetings usually lack thepersonnel necessary to discuss Level III topics, even if doingso would help the proposal effort.Level I FindingsEvents were Olassified as being at Level I if the mainobjective of the meeting was to draft text (words, visuals) orthe technical solution, activities which often happened intandem. As other researchers have noted, knowledge is builtby communities as they write together (Bruffee, 1986; Winsor,1990) . Examples of Level I events are Technicians 1 and 2245collaboratively building text during Event J (Down to Work),and Event Y (Coming Together) which finds a large group(Directors 1 and 2, Manager 1, Technicians 1 and 2) allengaged in checking text by reading and discussing it.A major finding in the building of the proposal at LevelI is the prevalence of the activity I have called recyclingtext, or using “clip-text.” Cerebellum writers refer often tothe use of “cut and paste” as a drafting practice, and myobservations show that drafters rely on text chunks frompreviously written proposals, technical reports, and lettersfrom vendors for assistance in many ways and at many stages ofthe text production process: in planning and building prose,in formatting, in using visual aids, and in dealing withnumbers. Although others have reported the prevalence ofusing boilerplates (Seizer, 1983) by individual, confidentwriters, my research shows collaborators who are oftenfrustrated by writing also using clip-text extensively. Inaddition, however, I conceive borrowing in a rather moreplastic fashion, as a process of drawing together ideas,experiences, and texts, of gathering knowledge andrepresenting it in text.Event A (The Starting Gun), at Level III, containsoverall process awareness comments which include writerspointing forward in the process to suggest Level I recyclingactivities. Manager 1 is directed (by Accountexec 1 atDirector 2’s suggestion) to seek out two previously writtenproposals as places to look for guidance. And again in Event246F (Under Consideration), at Level II, Director 2 looks forwardto Level I as he reiterates his belief in using alreadycomposed texts: “Get him [Director 1) to cite good qualityexamples and start cutting and pasting them.”The first event which shows collaborators building text,Event H (Problems, Problems), contains references to “cut andpaste.” Writers do not really cut; rather they copyappropriate sections from already complete documents andinsert them into the current document. Using writing as clipart is spoken of with some derision as yet another example ofAccountexec 1 trying, perhaps, to avoid the real work oforiginal composition. Technician 2 says in response toAccountexec l’s exaggerated offer to write the executivesummary: “Oh, yeah, he’ll just cut and paste.” Technician 1supports her in deriding Accountexec l’s proposition: “He’salready got a canned version.” In no other event, however,was clip-text referred to in anything other than positiveterms. Discussion of the use of clip-text visuals also occursin Event H (Problems, Problems) when Manager 1 asks: “Are yougoing to use diagrams?” and Technician 1 replies: “The sameones we used before—some minor modifications.”Collaborators (themselves cutting in and out of theprocess) reinforce the use of recycled ideas to and with eachother. They re-use ideas for the more effective drafting oftext, as later in Event H (Problems, Problems), whenTechnician 1 refers to his use of a “constant” as apresentation device borrowed from other piece of writing:247“Yeah, we did that before”. Such reinforcing also occurs whenAccountexec 1, Technicians 1 and 2 show Manager 1 what “style”(what might be called format) to use in the proposal. Manager1 leafs through a completed proposal and notices the use of“bullets” and “tables.” Even the RFP is consulted as a placefrom which format and phrases can be borrowed. We see Manager1, Technicians 1 and 2 consulting the RRP in Event P (Enterthe Dragons), so that their options in the proposal’stechnical solution will echo the requirements in the RFP.The degree to which the previous material can be useddepends of course on the experience of the writer. This iscertainly one reason why collaboration is so useful to newemployees working on a proposal. Sometimes, however, theBorden proposal writers find themselves struggling to resolvedifferent levels of knowledge, as in Event P (Enter theDragons) . Technicians 1 and 2 have not really used prepackaged text, but borrowed at least partially chunked piecesof information on ways to make a computer system available atclose to one hundred percent of the time. The choice of whichsolution to use must be made carefully, in this case becauseof cost.While clip-text must be chosen carefully and oftenmodified to work well, it can save time and help to target theaudience. Technician 1 and 2 quickly decide in Event J (Downto Work) on sections which can be extracted from previouswork, such as technical reports written for the same client,which can be inserted into the current proposal. When248Technician 2 expresses concern about getting material in forword-processing by a deadline date, Technician 1 replies that“It’s going to be tight.” She comments further, however, thatthey will be able to use “some cut and paste,” presumably as away to save time. Such assembling of already written text iseven more obvious in Event Q (The Recyclers), where what mightseem at first to be level-jumping (to Level II informationgathering) is really quintessential Level I drafting.Technicians 1 and 2 are more confident of their skill indealing with gathering and considering technical informationthan in their ability to draft.Text recycling by technicians, therefore, may be seen asan effort to recast their work in a more comfortable Level IIform; however, they are not the only writers to engage in suchan activity. Observations also suggest that Manager 1 andAccountexec 1 use clip-text, and Director 1 recommendsrecycling. All of these employees are executives comfortableat Level III. Director 1, who seems to be the one Cerebellumwriter most comfortable with original corttposition, discussesthe prevalence of clip-text in Cerebellum documents when hecomments to me how he often recognizes previously used phrasesin “new” documents. He further wonders how far back suchword-chunks could be traced in older documents.Collaborative Level I events are completely missingduring the early phase of text production. In fact, over thecourse of an eleven day proposal, drafting is not observeduntil Day Six, almost halfway through the allotted time.249Drafting then dominates middle and late events, as indeed itmust in order to meet the deadline. In total, there are tencollaborative Level I events discussed: four in the middlephase, five in the late phase, and one (Event P (Enter theDragons)) which serves as a bridge between the middle and latephases. Event P also marks the entrance of Director 2 as theobvious leader of the proposal effort at all levels (untilDirector 1 takes over the chief drafter role during Event TAsking the Tough Questions).Half of the ten events (Events J, M, Q, R, V) at Level Ishow collaborators primarily building text. Of the otherfour, three (Events K, T, Y) are primarily text checkingevents and two (Events H and P) reveal writers building andchecking fairly equally. Drafting the proposal textprogresses hand in hand with building the technical solution,despite Director l’s wishes to the contrary (see his finalcomment, Event A The Starting Gun). It is difficult tobelieve that, given the time pressures, any other approachwould be possible, however. Writing the words remindstechnicians of missing or contradictory parts of the approach,and discussing the approach directs them to write new sectionsof prose, or build a chart. Examples may be found in Event J(Down to Work) and Event V (Words, Words, Words) with thecomments of Technician 1, the technical architect in charge ofbuilding the solution and also drafting much of the technicalsection of the proposal. In Event V, Technician 1 debates theuse of terms and sentences with the comment: “Yeah. That’s too250vague. You do the proofing there, right?” The explanation isunclear, and so, it seems, may be their understanding of thetechnical steps which must be taken. Winsor (1990) noticedsimilar behaviour in the drafting of an engineer.In both cases, Technician 1 seeks collaboration for hisdecisions on the words and the solution. Much the sameprocess, building text and solution together, may also be seenin Event M (A Decisive Dyad) with Accountexec 1 and Manager 1.Here they discuss the proving procedure for the solution.Accountexec 1 suggests a technical procedure: “what we have todo is set up some sort of test” and Manager 1 respondsimmediately with a text idea: “so maybe [we could say] ‘Ourideas are flexible but one idea might be .. .‘“The review and discussion of text in events K, P, T, andY are especially important in view of the large number ofcollaborators working on the Borden proposal. There are sevenmajor contributors actually drafting text during the elevenday proposal effort (Directors 1 and 2, Manager 1,Accountexecs 1 and 2, Technicians 1 and 2) . Considerationmust also be given to the large number of other writerspresent through clip-text taken from the RFP, other proposals,letters, and reports. Text must be reviewed to check forcompletion, accuracy, effectiveness and also to smooth all ofthese voices into one voice to speak for Cerebellum throughthe complete Borden proposal.251Collaborators meet to ensure that they are graduallybuilding a unified document. “Rehearsal” seems to be one ofthe major collaborative activities of Level I, and it mostoften takes place during checking meetings. Writers rehearsetext by practising phrases with each other for possibleinclusion in the document.Such behaviour is first revealed in the data during LevelII Events B (Opening Moves) and F (Under Consideration).Although Events B and F are information sessions, during eachone we see writers looking ahead to Level I, when they will bedrafting text. In Event B, Accountexec 1 voices the wordinghe prefers: “But we will ‘minimize your risk, ‘“ even as hetries to inform Manager 1 about the client’s expectations. Hewants the proposal to use the phrase, and may be seen asoffering it to Manager 1 so it can be clipped and added to thedocument. Technician 2 uses the same technique of rehearsalin Event F, when she says to the group: “so we’re saying ‘putin a four port pad, because you have a choice ... so we’rerecommending four then you can have an option to upgrade it.’”We see such behaviour more frequently in Level I events.During the middle phase, three events reveal examples ofrehearsal from four different collaborators. For example, inEvent H (Problems, Problems), Manager 1 says: “We’ll say‘Here’s what you need to do.’” Rehearsal seems to serve achecking or previewing function in this event, and is alsoused here by Accountexec 1 to explain how he plans to word hissection of the proposal: “Well, not get into the252hardware/software ... [but] how they’re ‘looking for a hardwareplatform blah blah blah for efficiency of operation.’” EventM (A Decisive Dyad) also shows Accountexec 1 and Manager 1engaged in rehearsing text. Technicians 1 and 2 alsorehearse, perhaps as a way of making the document as clear aspossible by enuring that the headings match the accompanyingexplanations, as is shown in Event J (Down to Work).Late Level I events also show several examples ofrehearsal, but now they seem more to be a way of hurrying thegroup along. Director 2 dominates the rehearsal activities(except for one example by Manager 1 and one by Director 1)with comments in Events P (Enter the Dragons), R (A Woman’sPlace), and V (Words, Words, Words). Event P marks his entryas the leader of the proposal effort. He comments, during adiscussion of a spreadsheet, that, “We don’t want too manychoices for them to make. ‘If you want the absolute base, it’sthis.’” During Events R and V, Director 1 looks for quickconsensus on headings and titles. In this case, the presenceof a high status and extroverted personality seems to usurpthe rehearsal behaviour.Rehearsing ideas and exact words seems an ideal way towork with fellow collaborators. Writers are able to practicebefore hand, get confirmation or rejection on text in progressby giving words to ideas, keep the group acting in concertsince everyone can hear plans and adjust their own text, andhasten the decision process by getting input, which should253also cut down on the possibility that final reviews willreveal major problems.Other reviewing behaviours involve reporting on ways ofpresenting ideas. Event K (Covering Of f) for example, showsTechnician 1 checking his approach and the text he is writing(what he would call “covering off”) with Accountexec 1.Accountexec 1 drafts text as well, but is less effective attaking advantage of opportunities to check it, and perhapsthus misses the opportunity to merge his style with that ofother writers. This problem is particularly evident in EventP (Enter the Dragons), when Accountexec 1 submits a piece ofwriting and then leaves. The others, especially Technician 2,realize that the proposal has several styles. Technician 1comments, “Once everybody’s happy, then I think we just haveto do it in one style.” However, he will avoid being thewriter to bring about a unified voice, a role we see Director1 take late in the process. Indeed, trying to control the“versions” becomes a major job in the late phase of textproduction as we see especially in Event T (Asking the ToughQuestions) where Director 1 notes, “My temptation is to takeit away to a corner, but we’ve got three people working on it——we’ll get three different drafts.” The problem continuesduring the panic of the final day when writers struggle tokeep track of which sections of the proposal have been checkedand which have not.254As previously mentioned, seven writers collaborate atLevel I. During the ten events observed, Technician 1 appearseight times, and Technician 2 on seven occasions. Manager 1and Director 2 contribute during five events each.Accountexec 1 appears three times, Director 1 twice, andAccountexec 2 just once. Patterns of attendance arenoteworthy. Technicians 1 and 2 always appear together,except for one event (K) where Technician 1 collaborates withAccountexec 1 on his own. Technicians 1 and 2 work welltogether, and form a true partnership both in developing textand technical solution and in checking and justifying ideaswith others.All three of Accountexec l’s appearances occur in middlephase meetings. None of the directors appears at any middlephase meetings, but Director 1 or 2 or both are present at allbut one of the late phase meetings. Accountexec 2’s onlyappearance is also during the late phase. Manager 1 appearstwice in the middle phase, and three times in the late phase.It seems that the more senior, experienced personnel dominatelate phase Level I collaborations.Relationships among team members and the roles of theindividuals on teams change considerably during Level I eventsfrom the early through the middle to the late phase. As notedabove, Level I groups have more senior participants during thelate phase, and it is no surprise that the directors take overa leadership role, beginning with Director 2’s appearance inthe middle of Event P (Enter the Dragons). Much of Manager255l’s non-collaborative work, it seems, has been completedwithout significant input from Accountexec 1. My observationsof collaborative work show that his attendance at meetingsstops on Day Seven, and when Director 1 discusses thesituation with Manager 1, it is already Day Ten. Accountexec1 has missed two full days of collaboration (as corroboratedby Manager 1 in Event T (Asking the Tough Questions)).Accountexec 1 becomes a text reviewer, and Manager 1 adrafter. Directors 1 and 2 become the leaders of the textproduction effort.Other issues in team relationships are those surroundingTechnician 2. The relatively young and inexperienced femaleengineer begins to assert herself as an important team member.In early large group text-building sessions, she has been veryquiet, limiting her comments and taking very little part inmeetings. She supports Technician l’s ideas in Event H(Problems, Problems), especially in opposing Accountexec 1.However, Technician 2 works equitably with Technician 1 intheir own small group, expressing herself forcefully andeffectively. In Event J (Down to Work), for example, shecomments on a section heading: “That’s what’s confusing” andTechnician 1 readily agrees: “I think that’s the wrong name.”Her major challenge in the Borden proposal is to find herconfidence in larger sessions, where Technician 1 tries toavoid interrupting her (see for example, Event F UnderConsideration) but Director 2 proves more impatient--eveninsensitive until she finds a way to assert herself as she256does in Event R (A Woman’s Place). Here she manages to getand keep Director 2’s attention despite Technician l’s attemptto “assist” her by taking over a technical explanationhimself! By Event V (Words, Words, Words), Director 2 is evenasking for her input on drafting issues: “Instead of ‘setup’, can I say ‘configured’?”Several shifts in authority, based on drafting ability,occur during Level I. Accountexec 1 has his text completelyremoved from the proposal, and the sections are re-drafted byAccountexec 2, Manager 1, and Director 2. Manager 1 has hiswork severely criticized by Director 1, but will do some ofthe re-writing himself with Director 1 doing most of the redrafting. Director 2, Technicians 1 and 2 work well together,with Technicians 1 and 2 doing most of the original draftingand Director 2 reviewing. However, Director 2 also draftsadditions to technical sections and his work is reviewed byTechnicians 1 and 2. Tensions surrounding these changes areminimal except for one comment by Manager 1 (“I’ve done a lotthat I can’t use”) and the problems of Accountexec l’sbehaviour which are more obvious (see for example, Events Hand T). Much time is wasted on constant shifts in leadership,however, and much of the drafting takes place underconsiderable pressure during the final two days before thedeadline.Audience sensitivity, and the lack thereof, so evident achallenge at Levels III and II continues to be a problem ofarticulation and realization at Level I. Perhaps the257strongest example is in Event P (Enter the Dragons) whereManager 1, Technicians 1 and 2 struggle to get a unified ideaof the audience’s needs based on a close reading of the RFP.What is straightforward to the manager, however, is moreproblematic to the technicians, based perhaps on their greatertechnical knowledge of the problems and their previousexperience with the audience: “I don’t think he realizes whatit’s going to take—the dollar figure.”Obviously, there is a challenging tension here betweenknowing the audience well enough to be sure the proposalspecifically addresses their problems on the one hand, and notallowing a knowledge of audience eccentricities to overwhelmattempts to focus the presentation on the other hand. Thissubtle difference is handled by Director 1 in Event T (Askingthe Tough Questions), when he jokes about audiences and showsManager 1 a way to give them what they want AI’JD whatCerebellum wants. Director 2, in Event R (A Woman’s Place),also prefers a general perspective on audience which informsword choice, based on his skill with technology, and, perhaps,in dealing with audiences fearful of things they do notunderstand: “I don’t want to scare them that we’re doinganything significantly leading edge—’proof of concept’suggests something that’s never been done before, but this hasall been done before.”The schedule is maintained directly with comments onlyoccasionally during Level I. This may be because the draftingbegins so late in the process that writers are all too aware258that time is pressing. Event H (Problems, Problems), thefirst drafting meeting, shows examples of attempts to maintainthe schedule, and attempts to avoid that maintenance. Manager1 urges Accountexec 1 to take on drafting roles, whichAccountexec 1 avoids. Four days later, during Event T (Askingthe Tough Questions), Director 1 changes Accountexec l’s role,rather than leaving him out of the process, in order to gainhis experience with the client and maintain the schedule.Finally, Event Y (Coming Together) shows text checkinghindered by many writers trying to share one version of thedocument. Once again Director 1 suggests a change in approachbecause, “We’re not making the best use of everyone’s timehere.”Whereas Level III and II showed awareness of the rest ofthe writing process, Level I has almost no evidence of lookingforward, perhaps because drafting is the ultimate objective towhich the other levels have been pointing. One exception isbrief mention, made in Event M (A Decisive Dyad), to the oralpresentation of the proposal which may occur if Cerebellum isshort-listed. Manager 1 opens a conversation on the testingprocess which the proposal must outline by remindingAccountexec 1, “We’re going to be asked this.” Otherwise,writers do look ahead to upcoming drafting tasks, as in EventP (Enter the Dragons) where Technician 1 comments, “It’s abusy little section. We just have to round it out, polish itof f and make it clear in the document.”259Similarly, collaborators working at Level I do not engageto any great extent in level-jumping. Once again, timepressures seem to act as a motive to focus on Level I draftingactivities. One exception occurs in Event H (Problems,Problems) when Technician 1, still annoyed at the strategy hemust detail in words, attempts a jump from Level I to LevelIII: “It’s a weighted approach ... weighted to [Azure] .“ Heis quickly diverted away from Level III, however, by Manager1: “We discussed this and [Director 1] said: ‘Don’t stiranything up.’”In conclusion then, Level I events show collaboratorsstruggling to draft a proposal while still concerned overunresolved issues from Levels III and II. They seem to bewell trained to debate strategy, information and technicalsolutions. Choosing and using clip text, nurturing juniorteam members, learning to deal with problematic audiences andgenerally dealing much earlier with drafting problems mighthelp to make the writing process more efficient and moreeffective. Collaborative groups provide opportunities tolearn all of these skills, given the right team members andmore awareness of the issues. Then later, when some of thedrafting of the proposal is done alone, writers might carrywith them, into their offices, advice and ideas and approachesgained in concert with others.260CHAPTER NINE: IMPLICATIONS[un those communal endeavors whose goal issymbolic knowledge, the more we understand the waysymbols are used in the activity, the better we cancarry out that activity. (Bazerman, 1988, P. 317)As promised in Chapter One, my research has addressedsome gaps in knowledge about the collaborative writing processin the workplace. First, I have provided a detailedobservation of business writers at work (as urged by Ede andLunsford, 1990), and represented those observations in a modeldesigned to provide the necessary structure to organize andrender more schematic a complex process and focus furthercomments and research. Second, my observations have looked ata broad inclusive definition of writing, which embraces talk,the interactions surrounding planning and drafting, and themerging of authorship (encouraged by Couture and Rymer, 1989).Third, I have endeavoured to categorize various activitieswhich surround the creation of discourse. Most especially, Ihave looked at the relationship of one text to other texts (assuggested by LeFevre, 1987) in what I have termed the use of“clip-text.” Embedded within my main emphasis on levels ofengagement with text is the importance of each writer’srelationship with others as collaborators. I have definedcollaboration in one more setting and thus contributed to ourstore of knowledge about how writers work together.261General Implications for Advanced Literacy DevelopmentThe results of my research have general implications forthe development of better collaborative writing on the job andin the school.It is difficult to predict how Canadian businesses willbe operating in decades to come. Communication skills,however, will continue to form the foundation of competitiveand co-operative ventures in the future. Therefore, it isimperative that business communicators have the opportunity toimprove their ability to read and write, speak and listen inorder to further their own development and meet the needs oftheir company. Well-trained writers not only have valuableand transferable job skills, but are more able to assertthemselves and influence any company’s culture. If it is truethat communities define themselves by the discourse projectsthey adopt, then employees who learn better writing approachescan more fully participate in the life of the community. Thiscan only be accomplished by a continuing emphasis on literacyeducation.While educators frequently feel harassed by what oftenseem to be the demands of corporate Canada for workers whowill contribute to the realization of profits, perhaps it istime for schools to stop resisting or ignoring the concerns ofbusiness. Rather, we might look at our goals for writers,especially in the high school. There can be no doubt thatthere are differences between school writing and businesswriting. Our goals and therefore our approaches vary. The262primary goal of business writing is to realize financial gain,and attention to process, therefore, centres mainly on ways toimprove writers’ approaches so that they can improve theproduct and its ability to please a client.The main goal of the schools on the other hand is to helpall students to learn. Any given collaborative writingproject may be used to teach skills in such areas as grouprelationships, composition, and problem-solving. Approachesfocus on all students having an equal chance to participate,regardless of their skills, in order to develop abilitiesgradually. Business needs to understand where our goals mustdiffer. Students must be allowed to learn from mistakes thatwould get an employee fired! Educators could, however, offersome opportunities for students to work in groups asbusinesses might organize them, with time provided to discussthe effects of hierarchies, for example, on the group process.One of the main goals of my research, then, is to build abridge between business and the academy to improve thepossibility of future research completed together. Suchconnections need to be maintained, and more work needs to bedone in sharing objectives, procedures and problems inendeavouring to work in concert. Looking at two proposals inone company, even in the context of a two-year project in thisoffice, still gave me only a glimpse at what might be possiblewith more time spent working together on questions of mutualinterest.263Implications for Reading, Writing and Using TextsSince Cerebellum writers rely on clip-text strategies toa great extent, work also needs to be done to investigate thebest ways for employees and students to use other texts.Choice of sources is obviously crucial, but so is developingcriteria to help writers decide what to borrow (just a formator a whole section of a proposal) and how to integrate it intothe current proposal (with or without modification, forexample) . These recycling skills are sometimes not positivelyacknowledged in professional literature (Wallace, 1994) andneed more investigation.Ways to merge the resulting patchwork of text needsattention as well, perhaps by having less experiencedemployees work more closely with more skilled employees. Suchskill could also of course help writers to deal with textcontributed by different writers currently producing originaltext or contributing their own clip-text. Cerebellum writersshow confidence at strategizing and information gathering, butare far comfortable as flexible composers. Talented drafters,such as Director 1, find themselves left with much of the workin presenting ideas effectively during the late phase ofwriting. In the Borden proposal, problems with organizationand style in the written work were not discovered until thepenultimate day, which left very little time for re-drafting.As also advised by Bizzell (1982), training in discourseanalysis (for example, recognizing active and passive sentence264structures) could help writers to develop and then influencethe corporate writing style of their own community.Reading and writing are complementary communicative acts.My research shows that a lack of careful reading, crucial tothe use of other texts, resulted in difficulties and delaysfor at least fQur writers. The Request for Proposal (RFP)issued by Borden was read and debated by Manager 1 andAccountexec 1, and subsequently interpreted. Theirdiscussion, however, reveals a lack of careful attention tothe text, in analyzing the rhetorical problem (noted also asan important skill by Flower, 1989b). Rather than re-readingto understand the client’s desires, Accountexec 1 movesquickly to an interpretation which would minimize Cerebellum’sexposure to risk, without adequate consideration of otherpossibilities. Further problems are experienced byTechnicians 1 and 2 in writing the proposal as a response tothe RFP document, and by Manager 1 in following the formatsuggested by the RFP. More training in ways to interprettexts, such as looking for patterns of organization andcreating an outline, would help both business employees andstudents to gain and use valuable information from documents.Given the predilection of business communicators toborrow extensively from previously written documents, teachersneed to examine if and how such a skill might be taught in theschools. It is difficult to escape the similarity of usingclip-text to engaging in plagiarism, a much scorned practicewhich is cause for failure or even expulsion in many schools,265and yet remains a common student habit. How ironic, evendisconcerting then, to find that adult business writersroutinely employ such “borrowing” of text chunks, both wordsand diagrams, in composition tasks. It seemsa naturaloutcome of using other documents as models for format andstyle, an exercise used also by teachers to help students.And yet, where business writers progress to harvesting otherdocuments for phrases, paragraphs, and even whole sections toinsert in current texts, students are forbidden to engage insuch practices.In my own teaching experience, I was once berating acompetent Grade Eleven writer, Rod, for lifting large chunksof text out of an encyclopaedia and inserting them into hisreport. When asked why he would do such a thing, he replied:“Why not? I couldn’t say it any better.” When adjured toremember the rules of citation, he shrugged his shoulders. Hewas disinclined to use such a ponderous system of footnotesand references when he just wanted to get through his workquickly and efficiently. If he was sorry at all, it wasbecause he had been caught. At Cerebellum, his ability withclip-text would have been lauded, and herein lies the problem.What are the implications for us within the school system?One view would no doubt be that students like Rod need tolearn to write original compositions and they will not do soif allowed or encouraged to use clip-text. Business writers,after all, have different objectives. They draw on a corpusof discourse, owned by the company, in order to compete266efficiently in winning business contracts. In this senseperhaps, they borrow from themselves in much the same wayacademics do in writing articles which refer to previousresearch done by the self-same author. One major issue isthat of text ownership. We want students to understandproprietary authorship as it pertains in differentcommunities. Perhaps, then, students could be shown how todevelop one paper which builds on a previous paper, using someof the same sentences, paragraphs and diagrams, but alsoadding new material. Or students could work on essays ontheir own and then do group projects which borrow materialfrom these previous pieces of writing.The alternative is to proceed as we are now, ignoring theinclinations of students and the realities of business writingpractices. I believe it is better to consider the work àf theclassroom, at least from time to time, as a body of discourseand show students how to use each other’s work as sources oftext. The integration of text is, after all, by no means asimple matter, requiring new skills suchas merging of styleby the final writer. I am not advocating the unacknowledgeduse of words and ideas by students. Such academic behavioursneed to be learned as the habits of one discourse communitywhich students may wish to enter. However, other ways ofusing text sanctioned by other communities also need to beacknowledged.267Implications for Representing Audience and CommunityCerebellum is a community with varying degrees ofunderstanding about developing relationships among communitymembers, within proposal teams of employees, with otherbusinesses and with clients. Some collaborators are alreadyquite adept at including potential audiences as communitymembers, for example, Accountexec 2 on the Borden proposal andTechnician 4 on the Sundial proposal. Both of these employeesindicate a willingness to establish and maintain relationshipswith clients which focus on meeting their needs (see alsoBeck, 1992 for similar advice) by involving them in theproposal process. The experience of Technician 1, who alsovalues his relationships with vendors, indicates what harm maybe done when a collaborator cannot interact honestly withother businesses (for advice on this issue, see Bryan, 1992).Time needs to be spent in discussing varying approaches to theclient as an audience. In that Technician 1 is frequentlycritical of Cerebellum’s approaches, his experience could beexamined by students as an example of how disagreement mightproceed (Harris, 1989).Various behaviours enhance or impede collaboration: teamrelationships and audience sensitivity most clearly relate tothe ability of the community to expand and include audiencesas members. Disagreement also contributes to effectivecollaboration, suggesting that group members should be taughtwhen and how to introduce opposing points of view. Both theBorden and Sundial proposals show disagreement at Level III268benefiting the invention process. At other levels, conflictis usually not tolerated, and is often seen as a challenge toauthority or a threat to schedule maintenance. Technician 2finds she must persist strongly to gain a place on the team,which under more nurturing conditions would be encouraged bymore senior team members. Both assertive action and listeningto others need to be seen and encouraged as important skills(see also Lay, 1992; Tebeaux, 1990; Lunsford and Ede, 1990).Total Quality Control (TQC) principles (Deming, 1988;Goldman, 1993) suggest the characteristics of the bestcommunication situations. Many of these qualities mesh wellwith theories about the primacy of the social context to thecommunicative act currently favoured by the rhetoriccommunity. In order to foster a social environment whichsupports writers’ attempts to work effectively together,traditional barriers must begin to fall. These barriers maybe physical ones, such as separation of writers by officespace, or social ones, such as a counter-productive hierarchy.If communication is to be “decentralized, vertical-upward,interdepartmental, interdependent, trusting, long-term, group-oriented, reciprocal, immediate, nurturing feedback, flexible,and characterizedby close proxemics” (Goldman, 1993, p. 29),then the current approach to writing together must be analyzedand steps taken to encourage a more supportive writingatmosphere. Managers need skills as coaches in helping teammembers to work together at each phase of the writing process(Zaslow, 1991)269Too much writing in the schools is for teacher-as-examiner (Britton, 1970). Although teachers have sought togive students other audiences, often these writing tasks arethinly veiled tests which fool no one. If students are askedto “Write a letter to a business client” which they know willnever be posted, but rather evaluated by a teacher, then verylittle is being done to advance the ability or motivation ofstudents to write for real audiences. Involving the audience,as I discovered in my research, is difficult even for businesswriters who face such tasks every day. Students need to writefor a variety of audiences and for a variety of real lifepurposes, both on their own and with other writers. Althoughthis is a rather familiar refrain in education, moreconnection between schools and business would make possiblementoring situations, in person or via e-mail, where studentscontribute to the writing of projects for which audiences arecarefully realized. Students need to experience the powerfuleffect of a group coming together to adopt a discourse projectand see it through to fruition.The ubiquitous and frequently problematic group project,however, which might be assigned to give students practice indebating audience representation, needs to be re-examined.Although my research shows that writing together in businessis not without its problems, it persists as an approachbecause of time pressures and the need for various specialiststo contribute to large writing tasks where high quality isimportant. Students need training in better ways of working270together by building team relationships, identifying andinvolving an audience, maintaining a schedule, looking forwardto upcoming writing tasks and focusing on different levels ofengagement with their text.My research shows business writers writingcollaboratively in both hierarchical and non-hierarchicalgroups. Leadership and authority change according to meetingobjectives, resting sometimes with one person, sometimes withanother. Although status as a senior executive still commandspower in business, there are signs in my research that changeis possible. More equal, or what Ede and Lunsford (1990)might call “dialogic”, groups are present in business andmight serve as models for schools to imitate. Also, a wayneeds to be found for the exceptionally bright, extrovertedindividual to learn to contribute to a group effectivelywithout dominating inappropriately (Veiga, 1991). Thissuggests that the best students need a chance to worktogether, as top executives do at Level III, and that theyalso have time to work on their own, using other documents ascollaborative sources.Implications for Further ResearchIn studying collaborative writing using ethnographicmethods to provide the detailed description needed to developa model, I have begun a research path which I would hope tofollow myself and encourage others to join. Although I wasprevented from sharing much of my video-taped data due to271proprietary concerns expressed by Cerebellum, I was stillable to gather much more detail of actual writing behavioursby using video-tapes than would have been possible usingfieldnotes or audio-tapes, or by engaging in surveys orinterviews. The model I developed was intended to respond tothe needs of business writers for a concise, pragmaticrepresentation.Much more might be done, however. As my data becomesless sensitive with the passage of time, it might be digitizedand organized in non-linear format (on video-disc, forexample) for use by teachers, business writers and researchersto investigate issues beyond what I have attempted in my ownresearch. In addition, of course, using ethnographicapproaches and then developing models could also be used tostudy other collaborative writers in both business and schoolsettings as a way to gather and study data, representfindings, and establish relationships across various cultures.My research indicates that collaboration takes place atwhat can be roughly described as three levels of textualinvolvement. Each level has its own objective which isrealized by more or less unique activities. The basis forauthority shifts from level to level, as does the tone ofmeetings. Business writers need to observe, analyze anddiscuss the descriptions of collaboration in order to becomemore aware of approaches which may be helpful in developingproposals (as advised as well by Mclsaac and Aschauer, 1990).For example, the egalitarian spirit of Level III strategizing272may work well in text checking meetings during Level Idrafting. As well, non-hierarchical collaborations, such asthose shown by Technicians 1 and 2, could be highlighted asmodels (as in Ede and Lunsford’s 1990 dialogic mode) for otherpairs or small groups learning to work together. Hierarchiesbased on status (see also Cross, 1990) could be re-ordered sothat more experienced collaborators might assist the lessexperienced to learn team skills.Such a representation as my model of levels of engagementwith text needs to be investigated in other settings and withother kinds of documents with various groups of collaborativewriters. A logical next step for an educational researcherwould be to take video cameras into a high school and filmgroup writing tasks. Researchers could then examine this casefor similarities and differences with the Borden proposalprocess, and look at the model to suggest changes to it, or tosuggest new ways for schools to organize and teach groupwriting skills.In future, I would spend more time looking for ways toinvolve informants more directly in the research effort. Timepressures at Cerebellum mitigated against any but the mostfleeting conversations on my research. I would also like tomodify my methodology to include more time/record keeping byinformants of total time spent working on the project. Thiswould allow me to examine what portion of time I had captured,and what portions of time informants worked face to face withcollaborators as opposed to more private activities. Finally,273I would like to find some way to make the research processseem more immediately valuable to business writers, and thusperhaps more deserving of their attention.Finding common ground, working collaboratively, seems tome to be a valuable goal for future research with businesswriters and students. The benefits touch on a number ofissues common to education and business, most cruciallyinsights into, and the development of, advanced literacy.274BibliographyAllen, N., Atkinson, D., Morgan, M., Moore, T., and Snow, C.(1987). What experienced collaborators say aboutcollaborative writing. Journal of Business and TechnicalCommunication, 1(2), 70-90.Anderson, G. L. (1989). Critical ethnography in education:origins, current status and new directions. Review ofEducational Research, 59(1), 249-270.Anderson, K. and Jack, D. C. (1991). Learning to listen. In K.Anderson and ID. Jack, Women’s words (pp. 11-26). Londonand New York: Routledge.Anderson, P. (1985) . What survey research tells us. In L.Odell and ID. Goswami (Eds.), Writing in nonacademicsettings (pp. 3-83). New York: The Guilford Press.Aristotle (1932). The rhetoric of Aristotle CL. Cooper,Trans.). London: Prentice-Hall. (Original work circa 300B.C.)Asner, M. (1991). Creating a winning proposal. BusinessQuarterly, Winter, 36-38.Bazerman, C. (1988). Shaping written knowledge. Madison:University of Wisconsin Press.Beck, C. E. (1992). Toward a rhetoric of technical proposals:Ethos and audience analysis. Technical Communication,39(Feb.), 122-126.Bereiter, C. and Scardamalia, M. (1987). The psychology ofwritten composition. Hilisdale, NJ: Lawrence EribaumAssociates.Berkenkotter, C. (1991). Paradigm debates, turf wars, and theconduct of sociocognitive inquiry in composition. CollegeComposition and Communication, 42(2), 151-169.Best, J. W. (1981). Research in education (4th ed.). EnglewoodCliffs, NJ: Prentice-Hall.275Bizzell, P. (1982). Cognition, convention, and certainty:what we need to know about writing. Pre/text, 3(3), 213-243.Bizzell, P. (1987) . What is a ‘discourse community’? A paperpresented at the Penn State conference on rhetoric andcomposition.Britton, J. (1970). Language and learning. Harmondsworth,England: Penguin Books.Brodkey, L. (1987). Modernism and the scene(s) of writing.College English, 49(4), 396-418.Brown, R. L. and Herndl, C. G. (1986). An ethnographic studyof corporate writing: job status as reflected in writtentext. In B. Couture (Ed.), Functional approaches towriting: Research perspectives (pp. 11-25). London:Frances Pinter.Bruf fee, K. A. (1983). Writing and reading as collaborativeor social acts. In J. N. Hays, P. A. Roth, J. R. Ramsey,and R. D. Foulke (Eds.), The writer’s mind (pp. 159-169).Urbana, Illinois: National Council of Teachers ofEnglish.Bruf fee, K. A. (1984). Collaborative learning and the‘conversation of mankind’. College English, 46(7), 635-652.Bruf fee, K. A. (1986). Social construction, language, and theauthority of knowledge: a bibliographical essay. CollegeEnglish, 48(8), 773-790.Bryan, J. (1992). Down the slippery slope: Ethics and thetechnical writer as marketer. Technical CommunicationQuarterly, 1(1), 73-88.Burke, K. (1957). The philosophy of literary form. New York:Vintage Books.Collier, J. and Collier, M. (1986). Ethnographic film. In J.Collier, Jr. and M. Collier, Visual anthropology (pp.151-173) . New Mexico: University of New Mexico Press.276Cooper, L. (1932). The rhetoric of Aristotle. EnglewoodCliffs, New Jersey: Prentice-Hall.Corbett, E. P. J. (1989) . What classical rhetoric has tooffer the teacher and the student of business andprofessional writing. In M. Kogen (Ed.), Writing in thebusiness professions (pp. 65-72) . Urbana, Illinois:National Council of Teachers of English.Couture, B. and Rymer, J. (1989). Interactive writing on thejob: definitions and implications of collaboration. InM. Kogen (Ed.), Writing in the business professions (pp.73-96). Urbana, Illinois: National Council of Teachersof English.Creative Research Group. (1987) . Literacy in Canada: Aresearch report. Ottawa: Southam News.Cross, G. A. (1990) . A Bakhtinian exploration of factorsaffecting the collaborative writing of an executiveletter of an annual report. Research in the Teaching ofEnglish, 24(2), 173-202.Doheny-Farina, S. (1986). Writing in an emerging organization:An ethnographic study. Written Communication, 3(2), 158-185.Doheny-Farina, S. and Odell, L. (1985). Ethnographic researchon writing: Assumptions and methodology. In L. Odell andD. Goswami (Eds.), Writing in nonacademic settings (pp.503-535) . New York: The Guilford Press.Doheny-Farina, S. (1993) . Research as rhetoric: Confrontingthe methodological and ethical problems of research onwriting in nonacademic settings. In R. Spilka (Ed.),Writing in the workplace: New research perspectives (pp.253-267). Carbondale and Edwardsville: Southern IllinoisUniversity Press.Debs, M. B. (1991). Recent research on collaborative writingin industry. Technical Communication, 38(4), 476-484.Deming, W. E. (1988). Out of the crisis. Cambridge, Mass.:Massachusetts Institute of Technology.277Dewey, J. (1938). Experience and education. New York: CollierBooks.Doland, V. M. (1989). Hypermedia as an interpretive act. In V.M. Doland, Hypermedia, (pp. 6-19). Hypermedia (IEEE).Driskill, L. (1989) . Understanding the writing context inorganizations. In M. Kogen (Ed.), Writing in thebusiness professions (pp. 125-145). Urbana, Illinois:National Council of Teachers of English.Durst, R. K. (1990). The mongoose and the rat in compositionresearch: Insights from the RTE annotated bibliography.College Composition and Communication, 41(4), 393-408.Ede, L. (1984) . Audience: an introduction to research.College Composition and Curriculum, 35(2), 140-154.Ede, L. and Lunsford, A. (1984). Audience addressed/audienceinvoked. College Composition and Communication, 35(2),155-175.Ede, L. and Lunsford, A. (1985) . Let them write—together.The English Quarterly, 18(Winter), 119-127.Ede, L. and Lunsford, A. (1990). Singular texts/pluralauthors: Perspectives on collaborative writing.Carbondale and Edwardsville: Southern Illinois UniversityPress.Elbow, P. (1987). Closing my eyes as I speak: An argument forignoring audience. College English, 49(1), 50-69.Emig, J. (1971). The composing processes of twelfth graders.NCTE Research report no. 13. Urbana, Illinois: NationalCouncil of Teachers of English.Etter-Lewis, G. (1991) . Blackwomen’s life stories. In G.Etter-Lewis, Women’s words (pp. 43-58). London and NewYork: Routledge.278Faigley, L. (1985). Nonacademic writing: The socialperspective. In L. Odell and D. Goswami (Eds.), Writingin nonacademic settings (pp. 231-248). New York: TheGuilford Press.Faigley, L. and Miller, T. P. (1982). What we learn fromwriting on the job. College English, 44(6), 557-569.Farrell, T. (1976). Knowledge, consensus, and rhetoricaltheory. In J. L. Golden, G. F. Berguist, and W. E.Coleman (Eds.), The rhetoric of western thought (4thed.), (pp. 608-622). Dubuque, Iowa: Kendall/HuntPublishing.Flower, L. (1989a) . Cognition, context, and theory building.College Composition and Communication, 40(3), 282-311.Flower, L. (l989b) . Rhetorical problem-solving: Cognition andprofessional writing. In N. Kogen (Ed.), Writing in thebusiness professions (pp. 3-36) . Urbana, Illinois:National Council of Teachers of English.Flower, L. and Hayes, J. (1981). A cognitive process theory ofwriting. College Composition and Communication, 32(Dec.),365-387.Flower, L., Hayes, J. R., Carey, L., Schriver, K., andStratman, J. (1986) . Detection, diagnosis, and thestrategies of revision. College Composition andCommunication, 37(1), 16-55.Forman, J. (1991) . Collaborative business writing: A Burkeanperspective for future research. The Journal of BusinessCommunication, 28(3), 233-257.Forman, J. (1992). Introduction. In J. Forman (Ed.), Newvisions of collaborative writing (pp. xi-xxii).Portsmouth, Nm Boynton/Cook.Freed, R. C. and Broadhead, G. J. (1987). Discoursecommunities, sacred texts, and institutional norms.College Composition and Communication, 38(2), 154-165.Freedman, S.W., Dyson, A. H., Flower, L. and Chafe, W. (1987).Research in writing: past, present, and future. Technical279Report No. 1. Berkeley: Center for the Study of Writingat University of California, Berkeley and Carnegie MellonUniversity.Fulkerson, R. (1990). Composition theory in the eighties:Axiological consensus and paradigmatic diversity. CollegeComposition and Communication, 41(4), 409-429.Geertz, C. (1973) . Thick description: Toward an interpretivetheory of culture. In C. Geertz, Interpretation ofcultures (pp. 3-30). New York: Basic Books.Geertz, C. (1983). Local knowledge: Further essays ininterpretive anthropology. New York: Basic Books.Geertz, C. (1988). Works and lives: The anthropologist asauthor. Stanford, California: Stanford University Press.Gieselman, R. ID. (1982) . Reading, writing, and research:pedagogical implications. Journal of BusinessCommunication, 19(4), 23-38.Goldman, A. (1993) . Implications of Japanese total qualitycontrol for western organizations: dimensions of anintercultural hybrid. The Journal of BusinessCommunication, 30(1), 29-47.Goidman-Segall, R. (1989). Thick descriptions: a language forarticulating ethnographic media technology. Unpublishedmanuscript. Massachusetts Institute of Technology,Boston.Goidman-Segall, R. (1991) . A multimedia research tool forethnographic investigation. In I. Harel (Ed.)Constructionism (pp. 467-496). Norwood, NJ: AblexPublisher.Grant-Davie, K. (1992). Coding data: Issues of validity,reliability, and interpretation. In G. Kirsch and P. A.Sullivan (Eds.), Methods and methodology in compositionresearch (pp. 153-171). Carbondale and Edwardsville:Southern Illinois University Press.280Haas, C. (1989): How the writing medium shapes the writingprocess: Effects of word processing on planning. Researchin the Teaching of English, 23(2), 181-207.Halliday, M. A. K. (1969). Relevant models of language. In A.M. Wilkinson (Ed.), The state of language (pp. 26-37).Birmingham: The University of Birmingham School ofEducation.Hammersley, M. and Atkinson, P. (1990). What is ethnography?In H. Hammersley and P. Atkinson, Ethnography: Principlesin practice (pp. 3-30). London and New York: Routledge.Hampton, R. E. (1990). The rhetorical and metaphorical natureof graphics and visual schemata. Rhetoric SocietyQuarterly, 20, 347-356.Harcourt, J. (1990). How to write winning proposals. Trainingand Development Journal, September, 29-33.Harris, J. (1989). The idea of community in the study ofwriting. College Composition and Communication, 40(1),11-22.Herndl, C. G. (1991). Writing ethnography: Representation,rhetoric, and institutional practices. College English,53(3), 320-332.Herrington, A. J. (1989). The first twenty years of Researchin the Teaching of English and the growth of a researchcommunity in composition studies. Research in theTeaching of English, 23(2), 117-137.Howe, K. and Eisenhart, M. (1990). Standards for qualitative(and quantitative) research: A prolegomenon. EducationalResearcher, 1.9(4), 2-9.Jacob, E. (1987) . Qualitative research traditions: A review.Review of Educational Research, 57(1), 1-50.Jacobi, M. J. (1990). Using the enthymeme to emphasize ethicsin professional writing courses. The Journal of BusinessCommunication, 27(3), 273-292.281Jones, D. (1973) . Cultural fatigue. Anthropological Quarterly,46(1), 30-37.Kirsch, G. (1991). writing up and down the social ladder: Astudy of experienced writers composing for contrastingaudiences. Research in the Teaching of English, 25(1),33-53.Kleine, M. (1990). Beyond triangulation: Ethnography, writingand rhetoric. Journal of Advanced Composition, 10(1),117-125.Kroll, B. M. (1984). Writing for readers: Three perspectiveson audience. College Composition and Communication,35(2), 172-185.Kuhn, T. (1970) . The structure of scientific revolutions(second edition) . Chicago: University of Chicago Press.LaDuc, L. (1991) . Infusing practical wisdom into persuasiveperformance: Hermeneutics and the teaching of salesproposal writing. Technical Writing and Communication,21(2), 155-164.Langer, J. A. (1985). Musings. . .a sociocognitive view oflanguage learning. Research in the Teaching of English,19(4), 325-326.Lather, P. (1986) . Issues of validity in openly ideologicalresearch: Between a rock and a soft place. Interchange,17(4), 63-84.Lather, P. (1991). Feminist perspectives on empoweringresearch methodologies. In Getting smart: Feministresearch and pedagogy with/in the postmodern. NewYork/London: Rout ledge.Lay, M. M. (1982). The androgynous collaborator: The impact ofgender studies on collaboration. In J. Forman (Ed.), Newvisions of collaborative writing, (pp. 82-104).Portsmouth, NH: Boynton/Cook.LeFevre, K. B. (1987). Invention as a social act. Carbondaleand Edwardsville: Southern Illinois Press.282Lincoln, Y. S. and Guba, E. G. (1988). Criteria for assessingnaturalistic inquiries as reports. A paper prepared forpresentation at the Annual Meeting of the AmericanEducational Research Association, New Orleans, LA.Lunsford, A. and Ede, L. (1990). Rhetoric in a new key: Womenand collaboration. Rhetoric Review, 8(2), 234-241.Lunsford, A. (1992). Intellectual property, concepts ofselfhood, and the teaching of writing. Journal of BasicWriting, 11(2), 61-73.Lyon, A. (1992). Re-presenting communities: Teachingturbulence. Rhetoric Review, 10(2), 279-290.MacDougall, D. (1975). Beyond observational cinema. In D.MacDougall, Principles of visual anthropology (pp. 109-124). Hawthorne, NY: Aldine de Gruyter.Malone, E. L. (1991) . Facilitating groups through selectiveparticipation: An example of collaboration from NASA,(pp. 109-120). In M.M. Lay and W.M. Karis (Eds.),Collaborative writing in industry: Investigations intheory and practice. Amityville, New York: BaywoodPublishing.Mao, L. (1989). Persuasion, cooperation and diversity ofrhetorics. Rhetorical Society Quarterly, 19(4), 131-142.Mclsaac, C. M. and Aschauer, M. A. (1990) . Proposal writing atAtherton Jordan, Inc.: An ethnographic study. ManagementCommunication Quarterly, 3(4), 527-560.Mead, M. (1975). Visual anthropology in a discipline of words.In Principles of Visual Anthropology by M. Mead, pp. 3-10. Hawthorne, NY: Aldine Publishers.Miller, C. R. and Selzer, J. (1985) . Special topics ofargument in engineering reports. In L. Odell and D.Goswami (Eds.), Writing in nonacademic settings (pp. 309-341). New York: The Guilford Press.283Moss, B. J. (1992). Ethnography and composition: Studyinglanguage at home. In G. Kirsch and P.A. Sullivan (Eds.),Methods and methodology in composition research (pp. 153-171). Carbondale and Edwardsville: Southern IllinoisUniversity Press.Neilsen, L. and Willinsky, J. (1990). The Learning ConnectionsProject: A School/Workplace Partnership. Unpublishedmanuscript. Halifax: Mount St. Vincent University.Newman, R. G. (1988). Collaborative writing with purpose andstyle. Personnel Journal, April, 37-38.Nystrand, M. (1982) . Rhetoric’s “audience” and linguistics“speech community”: Implications for understandingwriting, reading, and text. In M. Nystrand (Ed.), Whatwriters know (pp. 1-25). New York: The Academic Press.Nystrand, M. (1989) . A social-interactive model of writing.Written Communication, 6(1), 66-85.Odell, L. (1985) . Beyond the text: Relations between writingand social context. In L. Odell and D. Goswami (Eds.),Writing in nonacademic settings, (pp. 249-280). NewYork: The Guilford Press.Odell, L. and Goswami, 0. (1985). Preface. In L. Odell and 0.Goswami (Eds.), Writing in nonacademic settings, (pp.viixi). New York: The Guilford Press.Papert, 5. (1980). Mindstorms. New York: Basic Books.Pare, A. (1991). Ushering audience out: From oration toconversation. Textual Studies in Canada I. Kamloops,British Columbia: The University College of the Cariboo.Pemberton, M. A. (1993). Modeling theory and composing processmodels. College Composition and Communication, 44 (1), 40-58.Perelman, C. (1982). The Realm of Rhetoric. Notre Dame:University of Notre Dame Press.Peshkin, A. (1988). In search of subjectivity--one’s own.Educational Researcher, 17(7), 17-22.284Purves, A. C. and Purves, W. C. (1986) . Viewpoints: cultures,text models, and the activity of writing. Research inthe Teaching of English, 20(2), 174-197.Purves, A. C. (1992). Reflections on research and assessmentin written composition. Research in the Teaching ofEnglish, 26, 108-122.Reither, J. A. and Vipond, D. (1989). writing ascollaboration. College English, 51(8), 855-867.Rorty, R. (1979). Philosophy and the mirror of nature.Princeton: Princeton UP.Roth, R. G. (1987). The evolving audience: Alternatives toaudience accommodation. College Composition andCommunication, 38(1), 47-55.Seizer, J. (1983) . The composing processes of an engineer.College Composition and Communication, 34(2), 178-187.Seizer, J. (1992) . More meanings of audience. In S. P. Witte(Ed.), A rhetoric of doing: Essays on written discoursein honor of James L. Kinneavy (pp. 161-177). Illinois:Southern Illinois University.Smart, G. (1993). Genre as community invention: A centralbank’s response to its executives’ expectations asreaders. In R. Spilka (ed.), Writing in the workplace:New research perspectives (pp. 124-140). Carbondale andEdwardsviile: Southern Illinois University Press.Smudde, P. (1991) . A practical model of the document-development process. Technical Communication, 38(3),316-323.Spilka, R. (1993). Influencing workplace practice: A challengefor professional writing specialists in academia. In R.Spilka (Ed.), Writing in the workplace: New researchperspectives (pp. 124-140). Carbondale and Edwardsville:Southern Illinois University Press.Staw, B. M. and Ross, J. (1989). Understanding behaviour inescalation situations. Science, 246(Oct.13), 216-220.285Stohl, C. and Schell, S. E. (1991). A communication-basedmodel of a small-group dysfunction. ManagementCommunication Quarterly, 5(1), 90-110.Tebeaux, E. (1990). Toward an understanding of genderdifferences in written business communications: Asuggested perspective for future research. Journal ofBusiness and Technical Communication, 4(1), 23-43.Thralls, C. (1992) . Bakhtin, collaborative partners, andpublished discourse: A collaborative view of composing.In J. Forman (Ed.), New visions of collaborative writing(pp. 63-81). Portsmouth, NH: Boynton/Cook.Trimbur, J. (1989). Consensus and difference in collaborativelearning. College English, 51(6), 602-616.Varner, I. I. (1988). writing in groups. Journal of Educationfor Business, 63(6), 274-276.Veiga, J. F. (1991). The frequency of self-limiting behaviourin groups: A measure and an explanation. Human Relations,44(8), 877-895.Vygotsky, L. S. (1978). Mind in society. Cambridge,Massachusetts: Harvard University Press. (Original workcirca 1933-35)Wallace, ]D. L. (1994). Collaborative planning and transformingknowledge. The Journal of Business Communication, 31(1),41-60.Willinsky, J. (1990) . The construction of a crisis: Literacyin Canada. Canadian Journal of Education, 15(1), 1-15.Winsor, D. A. (1990). Engineering writing/writing engineers.College Composition and Communication, 41 (2), 58-70.Zaslow, R. (1991). Managers as writing coaches. Training andDevelopment, July, 61-64.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items