Rhetorical Moves in Scientific Proposal Writing: A - ResearchGate

Rhetorical Moves in Scientific Proposal Writing:

A Case Study from Biochemical Engineering

by

Brad Mehlenbacher

February, 1992

Submitted to Carnegie Mellon University, College of Humanities and Social Sciences inpartial fulfillment of requirements for the degree Doctor of Philosophy in Rhetoric andDocument Design.

Committee: Karen Schriver (Chair), Erwin Steinberg, David Kaufer

© Copyright 1992, Brad Mehlenbacher

1

Acknowledgments

Support from two sources allowed me to devote the better part of two years to the research

and writing of this dissertation—from a Digital Equipment Corporation/Carnegie Mellon

Fellowship and from the Engineering Design Research Center at Carnegie Mellon. As well,

the support of my committee members—Karen Schriver, Erwin Steinberg, and Dave

Kaufer—had a significant influence on my motivation to carry out the project and,

ultimately, on the final written product. Finally (and acknowledgment pages always seem

to beg our listing anyone we’ve ever known), I’d like to thank my family, Mom and Dad, for

continuing to ask me when I’m going to be finished, and Jeff and Barb and Emma, for not

asking me when I’m going to be finished. Special thanks to Mike Domach for his technical

expertise, guidance, and general irreverence; to Joseph Petraglia, especially, and to

Charlie Hill for their endless and (sometimes painfully) perceptive feedback on every

draft of every chapter; to Jim Palmer, Tom Duffy, and Maria Truschel, for making my first

two years at Carnegie Mellon as stimulating and as challenging as they were; to Jane

Kalbfleisch, for helping me survive the dissertation-writing process; and to Carole

Chaski, Nancy Penrose, Steve Katz, Mike Carter, and Carolyn Miller at North Carolina

State University, for sheltering me from committee work while I finished the final

chapters. Also, thanks to Elizabeth, Bill, and Gary and everyone at the Balcony for

infusing a little jazz into the dissertation-writing process.

2

Table of Contents

Abstract.............................................................................................................................5Chapter 1—Introduction.....................................................................................................7Chapter 2—Previous Research on Scientific Discourse.........................................................17

Characterizing Scientific Proposal Writers............................................................18A Description of Science Adapted from Cognitive Psychology......................19A Description of Science Adapted from Organizational Behavior................22A Description of Science Adapted from the Sociology of Science...................25

The Historical Dichotomy Between Science and Rhetoric........................................28The Contemporary Rhetorical Interest in Scientific Discourse..................................33

The Interaction Between Cognition and Context...........................................35The Interaction Between Process and Product...............................................36The Interaction Between Description and Prescription.................................38

Acknowledging the Role of the Proposal in Science..................................................42General Conclusions................................................................................................49

Chapter 3—Two Pilot Studies of Scientific Proposal Writing..............................................50A Talk-Aloud Study of Research Proposal Writing.................................................51

The Participants........................................................................................51The Task....................................................................................................51Classifying the Protocols............................................................................52

Issues That Emerged from the First Pilot Study.......................................................54Defining an Audience for the Research Proposal..........................................55Approaching and Organizing the Research Proposal...................................58Constructing a Professional Ethos................................................................61Focusing on the “Noise” that Enters Protocol Data.......................................62General Conclusions....................................................................................65

Bringing Context into the Study of Proposal Writing...............................................67The Participants........................................................................................67The Open-ended Interviews........................................................................67

Issues That Emerged from the Second Pilot Study....................................................68The Organizational Politics of Proposal Funding.........................................68The Interaction Between Proposals, Funding, and Research..........................71Differences Between Funding Sources..........................................................73The Continuum from Proposal Funding to Technology Transfer......................75General Conclusions....................................................................................78

Implications of the Pilot Studies for Research on Proposals.....................................79Chapter 4—A Case Study from Biochemical Engineering.....................................................83

Data Sources and Collection...................................................................................86A Participatory Design Approach to Data Collection..............................................90Data Coding and Analysis......................................................................................93Analysis of the Collaborative NIH Proposal..........................................................96Analysis of Raymond’s Initial Proposal Effort........................................................98Analysis of the Journal Article...............................................................................98Collected Iterations on Data Interpretations...........................................................99

3

Table of Contents, Continued

Chapter 5—Results of the Case Study.................................................................................101Raymond’s Background and Research Interests........................................................101Raymond’s Context for Writing...............................................................................104Raymond’s “Web of Writing”.................................................................................107The Collaborative Proposal-Writing Project...........................................................120Raymond’s Initial Proposal-Writing Effort.............................................................125Raymond’s Earlier Journal-Article Project...............................................................128Managing the Collaborative Proposal Project..........................................................134

The Fourteen Proposal Drafts.....................................................................134Writing and Re-writing the Proposal’s Specific Aims Section......................138Exchanging Notes to Aid Collaboration......................................................141The Two Taped Meetings............................................................................145

Limitations of the Case Study................................................................................148General Conclusions and Remarks...........................................................................151

Chapter 6—Discussion and Implications.............................................................................155Methods as Windows on Proposal Writing..............................................................161Scientific Proposal Writing as Storytelling.............................................................163Scientists and Engineers as Rhetors.........................................................................173Implications for Research and Teaching..................................................................179Appendix A—Schriver’s Agenda of Research Questions..........................................183Appendix B—Instructions for Taping a Protocol.......................................................185Appendix C—Questions Asked of Fifteen Academic Researchers.............................186Appendix D—The Episodes and the Multiple Data Types.......................................187Appendix E—Questions Asked of the Biochemical Engineer....................................188Appendix F—Chronology of Raymond’s Writing Projects.........................................189Appendix G—Results of the Data Analysis.............................................................190Appendix H—Chronology of the 14 Proposal Drafts................................................206Appendix I—Evolution of the NIH Proposal’s Specific Aims Section.......................207Appendix J—Evolution of the Proposal’s Three Major Sections.................................222

Bibliography.....................................................................................................................224Vitae.................................................................................................................................257

4

Abstract

Rhetorical Moves in Scientific Proposal Writing:

A Case Study from Biochemical Engineering

Brad Mehlenbacher

Proposal writing in the sciences and engineering has only recently

received attention by researchers interested in rhetoric and writing. In

this dissertation, two pilot studies are introduced. The first is based on

fifteen talk-aloud protocols of professional writing students and, the

second, on fifteen open-ended interviews with academic researchers from

various fields. Both studies, while offering insights into the nature of

proposal writing, raise important issues that require further

investigation. In particular, the studies reveal the need for research that

describes the following: (1) how goals, intentions, and plans interact with

contextual constraints and opportunities; (2) how writing processes and

written products are produced over time, and; (3) how descriptions of the

proposal-writing process can be used to inform current scientific and

technical writing pedagogy.

A third study aimed at addressing these issues is then described.

This study, of a biochemical engineer writing research proposals, extends

over two years. Based on multiple data-collection techniques, the study

expands our current understanding of the proposal-writing process. It

reveals that proposal writing, journal writing, and laboratory activities

5

are interdependent. Proposal writing in biochemical engineering is a

dynamic process in which academic researchers negotiate numerous

constraints—between their characterizations of the proposal’s perceived

audience and that audience’s reaction to their text, between the proposal

format and discourse conventions of the field, and between their texts,

research goals, and the goals of their collaborators. Proposal writing is

not, as some researchers have suggested, an activity that takes place

prior to scientific research and publication. Instead, in writing a

proposal, scientists must make numerous rhetorical choices regarding the

presentation of existing data, of their research experience, and of their

research intentions.

Importantly, applying different methodological approaches to

the study of writing in biochemical engineering presents different

“windows” on the proposal-writing process. Finally, I argue that

rhetoricians and sociologists are guilty of “objectifying” scientific and

technical writers—that is, of treating them as non-reflexive and a-

rhetorical—when, in fact, we may gain a better understanding of their

sensitivity to rhetorical issues by studying their writing processes in

collaboration with them.

6

Chapter 1—Introduction

. . . now increasingly, scientists [have] to demonstrate their analyticskills in planning and writing extensive proposals for well-thought-outprojects. They [have] to demonstrate their mastery of relevant literatureand techniques of research as well as argue their way around currentcontroversies in the field (58).

Mukerji, C. (1989). A Fragile Power: Scientists and the State. Princeton,NJ: Princeton UP.

. . . the National Institutes of Health . . . during 1989 funded 27 percent ofthe research proposals it received. It’s estimated that percentage willdrop to 23 percent this year. In 1975, the NIH [National Institutes ofHealth] . . . provided funding at nearly twice that rate—45 percent of itsapplications (B1).

Niederberger, M. (1990). Science Research Funding Shrinks. ThePittsburgh Press. Sunday, December 9.

Rhetoricians, social scientists, biochemical engineers, whoever; we’re allbasically in the same business—trying to account for uncertainty.

Notes taken during a conversation with Raymond, a biochemicalengineer.

Traditionally, the competitive nature of science has been de-emphasized in favor

of a picture of science that emphasizes its consensus-building and contributory nature. The

scientific enterprise, however, as numerous researchers in rhetoric, sociology, psychology,

and philosophy have made clear to us, is very much a “business” in which scientists

negotiate and compete—sometimes with intense aggressiveness—for limited human,

technological, and monetary resources. Only when sociologists and rhetoricians of science

have highlighted the raging controversies underlying historical and contemporary

7

scientific investigations (cf., Markle & Peterson, 1981; Mazur, 1981; Nelkin, 1978, 1979,

1987) have we begun to realize that all is not as it seems in the Mertonian tower of scientific

inquiry. The popular characterization of scientists, as disinterested and objective

observers, many now regard as a myth; at its very heart, science is a social and political

activity. And the funding process is perhaps the most explicitly political aspect of all (cf.,

Cole, Rubin, & Cole, 1977; Gustafson, 1975; Lawrence, 1978; Mitroff & Chubin, 1979).

Given the key role that funding plays in the development and extension of science,

it is somewhat surprising that research proposal writing—the activity of producing the

formal documents that are at the center of the funding process—has only recently received

attention by rhetoricians and writing researchers. There are two major explanations that

might clarify why this is so. First, the study of scientific and technical discourse has been

traditionally outside the “proper” domain of rhetorical analysis (Kelso, 1980; Overington,

1977; Prelli, 1989a; Wander, 1976; Weaver, 1970; Weimer, 1977). However, contemporary

rhetoricians like Richard E. Young (1978, 1987) and Carolyn R. Miller (1985) have called

for a theory of scientific and technical discourse as argument, and this mandate has

heightened the interest in applying rhetorical analyses to scientific and technical texts.

The second explanation for why few researchers have studied scientific proposal

writing is that most researchers have limited their analyses to the most visible and central

type of writing that scientists do—journal-article writing. After all, it is this writing,

scientists would contend, that communicates new knowledge to the field and contributes to

consensus-building in science; and it is, therefore, the type of writing that rhetoricians and

sociologists interested in science have tended to privilege (e.g., Bazerman, 1988; Bernhardt,

8

1985; Campbell, 1973, 1975; Gilbert & Mulkay, 1980; Gragson & Selzer, 1990; Gross, 1985,

1990a, 1990b; Harmon, 1989; Rymer, 1988; Swales, 1984; Swales & Najjar, 1987).1

In 1985, Mark Haselkorn asserted that although “very little empirical research is

conducted in the area of proposal writing, there is just enough to show what proposal

writing research can do and how desperately it is needed” (273). Ironically, in that same

year, Greg Myers (1985b) produced one of the first of such analyses. His studies of the

writing processes of academic biologists (1985a, 1985b, 1990) revealed that the relationship

between academic writing for publication and proposal writing for funding were strongly

linked. In fact, Myers (1985b) found that both biologists incorporated in their proposals

“passages from . . . article manuscripts” that he had studied in the past (598). In discussing

what he called the complex “web of writing” that makes up scientific, technical, and

professional discourse,2 Myers (1990) pointed to the tension that his proposal writers

experienced “in their attempts to present their work as interesting” and yet simultaneously

“show that it is original and yet entirely in accordance with the existing discipline” (59).

Myers’ primary argument is that, although scientific articles are usually the

“central” focus of researchers interested in scientific discourse, in his opinion, the research

1 Moreover, it is interesting that every one of the analyses cited here emphasize the

introductions and, occasionally, the materials and methods sections of scientificjournal articles. The general consensus among these researchers is that scientistsconstruct a plausible story-line regarding their place in the literature, the potentialcontribution of their study, and the means by which other scientists can replicatetheir methodological procedures.

2 For the purposes of this dissertation, I have intentionally conflated the termsscientific, technical, and professional discourse; numerous other expressions areused to characterize studies of the workplace composing processes of professionals,including writing in the professions, writing in nonacademic settings, pragmaticwriting, on-the-job writing, and so on. What most of these studies have incommon, in Debs’ (1989) words, is that “they document . . . the importance ofcommunication . . . in scientific and technological organizations” (33). In short, Iam interested in discussing writing as it occurs in the professions and, specifically,the proposal-writing process. Henceforth, I will use the term “scientific writing”as a rubric for the above types of writing.

9

proposal represents the most “basic” form of scientific writing that scientists do. His

contention is that

. . . for many scientists heading large laboratories, proposals are in onepractical sense the most basic form of scientific writing: the researchersmust get money in the first place if they are to publish articles andpopularizations, participate in controversies, and be of interest tojournalists. For these researchers proposal writing is by no means anoccasional administrative duty; it is a constant effort that may involveapproaches to a number of different agencies, that may take about aquarter of the [scientist’s] working time, and that requires more and moreattention as grants are given for shorter periods, and fewer projects arefunded (41).

With the exception of Myers’ (1985b, 1990) studies, however, very little research

has examined the nature of scientific proposal writing, particularly its long-term

relationship with journal-article publishing, laboratory work and experimentation, and

the funding process in general. How do academic researchers characterize, for example, the

perceived audience of their research proposals? In turn, how does the perceived audience

for research proposals influence the organization and content of research proposals? How

does the proposal-writing process influence and alter scientific data-collection techniques

and laboratory practices? How do scientists negotiate between the tinkering and

“messiness” of actual laboratory practices and still adhere to rigorous scientific discourse

conventions (cf., Kaufer & Geisler, 1989)?

Bazerman (1988) has argued for the importance of studying the complex negotiation

“between individuals bidding to work on, modify, develop, elaborate, or apply part of [an

academic] theory and [the] employers, funders, editors, referees, critics, and audiences who

grant the researcher various powers to continue, publicize, and gain acceptance for their

work” (184). Figure 1 represents the different types of negotiation that face scientific

proposal writers. Because of my interest in technical, as well as scientific settings, I have

10

extended Bazerman’s (1988) model to include the goal of theoretical and practical

contributions to knowledge.

Scientificand

TechnicalResearchers

Bidding To

theory practice

elaborate work on

apply

enhanceextend

modify

develop

Existing

With the Intention of

extending publishing

continuingaccepting

In Negotiation Withemployers

funders

audiences review panelists referees

critics

editors

Figure 1: The complex interactions that face scientific and technical proposal writers;adapted from Bazerman (1988).

Figure 1 is deliberately designed to be read in two directions: either from the

scientific and technical researcher out or from the scientific community in. Thus, scientific

and technical researchers are constantly bidding to work on, elaborate, cultivate, extend,

fine tune, or apply existing theories and practices with the intention of extending,

continuing, publishing, or accepting existing theory or practices in negotiation with editors,

employers, audiences, funders, critics, review panelists, and referees. For example, two

dominant theories that biochemical engineers are currently interested in exploring are the

view of enzyme structures as being domain independent versus the view that enzyme

structures are heterogeneously organized; however, because the biochemical engineer that I

studied is interested in combining the two theories, he must make a painstaking argument

11

for the inadequacy of applying either theory in isolation (see Chapters 4 and 5). Or, in the

case of four of the academic researchers interviewed in Chapter 3, researchers can bid to

enhance current computer storage capabilities, an activity which is usually defined as

being “more applied” than theoretical in nature.

It is notable that Bazerman’s framework for future research on the construction of

scientific research and writing de-emphasizes the important role that technology and

instrumentation play in science. Perhaps this oversight is the result of the commonly held

belief that there are sharp differences between scientific goals (i.e., knowing that) and

technological goals (i.e., knowing how) (cf., Miller, 1985; Skolinowski, 1966). Miller, for

example, in her (1985) article “Invention in Technical and Scientific Discourse: A

Prospective Survey,” cites Hannay and McGinn’s (1980) argument that science is the

business of creating knowledge and technology is the business of creating products or

processes. I would argue, however, that the dichotomy between science and technology is,

as with all dichotomies, generally problematic. To say that scientists are motivated to

explain nature and that technologists are motivated to build tools is to de-emphasize the

critical interaction between equipment and observation. After all, in his discussion of

Compton’s classic research program, Bazerman (1988) notes that Compton was “constrained

by what mathematics, logic, and prior well-established theory allow one to say, by w h a t

available equipment can do, and by what data actually turned up” (200) [Italics added].

12

The dichotomy between scientific and technological practice has been further

blurred given the emergence of a significant number of sub-disciplines or specialty research

areas (e.g., applied physics, biochemical engineering, etc.), since it is generally difficult to

label their concerns as being strictly theoretical or strictly practical. Janich, in his (1978)

article “Physics—Natural Science or Technology?” argues convincingly that traditional

distinctions between scientific inquiry and technological “tinkering” are problematic since

scientific data are never collected without scientific machinery and since an understanding

of scientific machinery inevitably informs, influences, and constrains scientific knowledge

claims. He points out that “doing experiments is more an activity to produce technical

effects , which can be described appropriately as engineering rather than as a scientific

activity, properly speaking, as a construction of machines rather than as an inquiry into

nature, as an attempt to produce artificial processes or states rather than as a search for

true sentences” (11). To Janich, Scientific inquiry is therefore more accurately described as

“the technology of measurement and the technology of observation” (21).

Gökalp (1990), as well, has discussed the complex interaction between

instrumentation, experimentation, and theoretical research in science. He argues that “one

of the major influences of . . . new techniques has been to increase interactions between

experimental and theoretical lines of work at several levels—the work place, the

individual, and the cognitive structure of new research lines” (298). Given this perspective,

then, it seems probable that one of the reasons scientists are currently interested in

documenting the structure of quarks (rather than the structure of atoms) is because they

have access to equipment that allows them to “see” quarks.3

3 This observation has been made by numerous rhetoricians and sociologists

interested in the construction of scientific knowledge (Bloor, 1976; Campbell,1975; Collins, 1982; Knorr-Cetina, 1983), but it is also notable that scientistsand engineers (at least the ones that I have interviewed formally and informally) are

13

In addition to calling for the study of scientific and technical researchers,

Bazerman’s (1988) model emphasizes another crucial aspect of the bidding process in

science—that is, the audience or audiences that evaluate scientific bids. Price (1986) has

identified the complex formal and informal communication networks that operate in science

(particularly between scientists and their journal reading community), and I would extend

his discussion of scientific audiences to include funders, editors, referees, administrators,

peers, graduate students, colleagues, employers, and other potential audiences. Bids, then,

are “submitted” to these numerous audiences in a variety of forms, from informal

conversations in the laboratory hallway to formal experimental or theoretical articles sent

to refereed academic journals.

Finally, the goal or result of this bidding or process of negotiation is the extension,

publication, acceptance, or continuation of the proposed research, theory, or practical

application. This extension, of course, feeds directly into existing theory and practice and,

in turn, informs future theory and practice. Indeed, one of the key relationships that I

explore in this dissertation is the relationship between proposal writing and scientific and

technical research. As I pointed out earlier, the common characterization of the

relationship between scientific research and writing tends to privilege, first, scientific

research activities and, second, the scientific journal article. Where scientific research

activity is the central focus, scientific discourse is often cast as either the non-problematic

communication of scientific findings or as a deceptive mask that hides scientific activities.

And where the scientific journal article is the central focus, other types of scientific

well aware of the practical and theoretical strengths and weaknesses of theequipment they use; indeed, their formal discourse is structured to include, in thelimitations section, considerations of—not only reliability, generalizability, andvalidity—but also the limitations of their data-collection equipment.

14

discourse such as the laboratory notebooks, sketches, doodles, notes, or research proposals

written by scientists, are relegated to the background.

Proposals, however, are not simply documents that are produced “after the facts

are established or well worked out” (as one scientist I interviewed characterized his

experience writing proposals). Rather, the process of writing a research proposal—as with

any composing effort—must surely influence or inform the research being represented in the

proposal. Chapter 5 presents my findings in regard to the relationship between proposals

and scientific research.

Finally, adapting Bazerman’s model or framework for future research interests me

for two reasons. First, it suggests that science is both a cognitive and a social enterprise.

That is, scientific and technical researchers are placed at the heart of the model and are

surrounded by a myriad of social constraints and potential audiences. And second,

Bazerman’s model implies that we still have a great deal to learn about discourse in science

and points to the challenging task that rhetoricians interested in science and technology

have set for themselves.

Thus the need for an exploration of the research proposal as rhetorical negotiation,

an exploration which begins to answer some of the following important questions: How do

academic researchers plan, write, and revise proposals for research funding and how does

the process of proposal writing shape academic research activities? What types of

rhetorical moves do academic researchers engage in when they produce proposals for

research funding? How does the research proposal as form enable and constrain academic

discourse and knowledge? How do scientists characterize their audiences and how do those

notions of audience influence the writing of research proposals?

15

Chapter 2 presents a detailed literature review which draws on research from

cognitive science, academic and nonacademic writing research, the sociology and history of

science, and the rhetoric of scientific and technical discourse. A detailed argument is be

made for why scientific and technical writing, in general, and proposal writing,

specifically, have normally been de-emphasized as objects of inquiry by writing

researchers. In particular, two exemplary studies will provide the starting point for my

discussion of proposal writing in science and engineering—Greg Myers’ (1985b, 1990) “The

Social Construction of Two Biologists’ Proposals” and Jone Rymer’s (1988) “Scientific

Composing Processes: How Eminent Scientists Write Journal Articles.”

16

Chapter 2—Previous Research on Scientific Discourse

The pieces all fit together, and they . . . were fragments at the beginning,and what’s interesting is that each section is almost an independent unitin itself. . . . And so each section is almost like a microcosm of the entirepiece. So within each section I . . . [am] looking for order, fitting ittogether, and then I’m really not sure how the various pieces are going tofit together, so I work on the pieces . . . and then I’m going to put ittogether (interview with Subject-Scientist J, 226).

Rymer, J. (1988). Scientific Composing Processes: How Eminent ScientistsWrite Journal Articles. Writing in Academic Disciplines: Advances inWriting Research, Vol. 2. D. A. Jolliffe (Ed.). Norwood, NJ: Ablex, 211-250.

So we wrote it to the audience I described and, you know, wrote kind ofdefensively like you try and do in science. You see something. What areyou seeing? Is it real?

First open-ended interview with Raymond, a biochemical engineer.

In this chapter, I will begin by characterizing my object of inquiry, that is, the

proposal-writing scientist and his or her context for writing. To do so, I draw on literatures

from cognitive psychology, organizational behavior, and the sociology of science.4 I then

outline why the issues I am interested in have been ignored historically by rhetoricians

and writing researchers. This is followed by a review of numerous contemporary studies of

4 I recognize that, in separating the research and literature of these disciplines, I run

the risk of reducing or de-emphasizing the many issues they have in common.Gross (1990), as well, observes that it is getting more and more difficult toidentify traditional disciplinary boundaries: “Thirty years ago the humanisticdisciplines were more easily definable: historians of science shaped the primarysources into chronological patterns of events; philosophers of science analyzedscientific theories as systems of propositions; sociologists of science scrutinizedstatements aimed at group influence (Markus, 1987, 43). In the last too decades,however, the humanities have been subject to what Clifford Geertz has called ‘ablurring of genres.’ As a result, ‘the lines grouping scholars together intointellectual communities . . . are these days running at some highly eccentricangles’ (1983, 23-24).”

17

scientific and technical discourse emphasizing, in particular, the literatures from the

sociology, psychology, and rhetoric of science. Finally, I outline the need for a study of

proposal writing in science and set the stage for Chapter 3—the description of two pilot

studies of proposal writing and the research funding process.

Characterizing Scientific Proposal Writers

I want to begin my investigation by characterizing a proposal-writing researcher in

charge of numerous graduate students and a well-equipped laboratory in a university

setting. Not only is the researcher faced with a highly institutionalized, competitive

academic challenge, but he or she is also constantly in search of new and relevant sources of

research funding. And an integral part of the scientific funding process is the writing of

proposals for research funding. As Eaves (1984) asserts,

Whether referred to as “grantsmanship” or “researchmanship,” thescholarship required for a successful research-grant application is asdemanding as that for a lecture, a report for publication, or a textbook.Preparation of a grant application is a scholarly endeavor that combinesthe values of a scientist and the skills of a scholar: dedication,enthusiasm, standards of excellence, intellectual honesty, ethicality,disciplined thinking, and clear writing (151).

To complicate things further for the researcher, proposal writing in the academy is

no longer an individual endeavor; rather, it is a complex social process.5 Academic

researchers often interact in large organizational networks that might include, not only the

corporate and federal funding agencies themselves, but also their peers, students, and

academic funding representatives. What, then, are the exigencies facing the proposal-

writing researcher? How does the academic researcher plan and manage the collaborative

nature of the proposal-writing process? How does he or she characterize the intended

5 See LeFevre (1987) for an explication of the contemporary rhetorical view that

invention is a social, rather than an individual, endeavor.

18

audience or audiences for the proposal, especially given the increased competition for

federal and corporate funding (Kenward, 1984; Mandel, 1983)?6

A Description of Science Adapted from Cognitive Psychology

Cognitive psychology gives us an excellent starting point for any discussion of

scientific writing in context. The literature on problem-solving helps us describe the

researcher’s relationship with his or her environment. That is, any individual researcher

can be characterized as a symbol-making, symbol-using “system” working within a complex

environment where both the system and the environment are informed by one another.

Thus, the researcher operates as a problem solver, that is, he or she attempts to

discover—through varying combinations of trial, error, and selectivity—accurate state

descriptions and process descriptions of some element of nature (Newell & Simon, 1972).

Moreover, the researcher’s problem space is an ill-structured one, consisting of complex and

changing problems, goals, sub-goals, and the researcher’s existing knowledge of the

solution constraints (cf., Simon, 1979; Voss, Greene, Post, & Penner, 1983).

In addition, as with scientific activity in general, the production of scientific

discourse is an ill-structured problem, that is, it offers no single solution path, no

checkmate (Kotovsky, Hayes, & Simon, 1985; Pennington, 1985; Spiro, Vispoel, Schmitz,

Samarapungavan, & Boerger, 1987). It is what Chi, Glaser, and Rees (1982) call a “real

world problem” and, as such, “presents new obstacles that were not encountered previously

6 The epigram from the Pittsburgh Press that introduces chapter 1, for example,

estimates that the NIH funded 23 percent of its applicants in 1990, a drop from 45percent in 1975 (cf., Novello, 1985, who estimates that approximately 50 percentof all grants received by the NIH in 1972 were funded, as compared to 37.3 percentin 1984, 281). De Bakey (1976), too, has pointed out that the NIH funded asmuch as 50 to 60 percent of the applications it received in 1976 (5). And Mitroffand Chubin (1979), referring to the National Science Foundation (NSF), cite thegrowing interest in scientific peer review as stemming, in part, from the newcompetitive funding situation (199).

19

in puzzle-like problems” since “the exact operators to be used are usually not given, the

goal state is sometimes not well defined” and “a large knowledge space” is essential (7).

That is, the agenda and goals of a productive academic researcher inevitably evolve as he

or she continues to collect and integrate new information over time.

Our proposal writer, then, is embedded in a complex task environment and faces

multiple exigencies—he or she must contend with methodological constraints, conflicting

representations of scientific and technical phenomena, existing research opportunities and

trends, graduate student and laboratory management, and so on. Simultaneously, our

researcher must attempt to construct his or her audience’s criteria for success, “why would

the NIH want to fund my research?”, and emphasize or de-emphasize research issues

according to the interests of the faculty, college, and academic discourse community in

general. In Bazerman’s (1988) words, “. . . even while the literature, research program,

problem formulation, experimental design, and data constrain the solution’s formulation,

all these earlier constraints are presented in the context of a formulation of the world that

takes the findings for granted” (202).

Like the literature on problem solving in cognitive psychology, the literature on

scientific discovery processes provides writing researchers with a vocabulary for

approaching scientific and technical invention, heuristics, and problem solving (e.g.,

Mansfield & Busse, 1981). Specifically, Simon and his colleagues have done numerous

studies of scientific discovery procedures (Bhaskar & Simon, 1977; Bradshaw, Langley, &

Simon, 1983; Kulkarni & Simon, 1988; Langley, Simon, Bradshaw, & Zytkow, 1987;

Langley, Zytkow, Simon, & Bradshaw, 1983). Their findings describe how scientists

represent data, develop hypotheses, design experiments, and generate new research

20

problems. Their findings, however, de-emphasize the role that texts and, in particular,

proposals for research funding, play in scientific communities.

In their study of Hans Krebs’ research on glutamine synthesis, for example,

Kulkarni and Simon (1988) argue that most scientific problem solving is guided by general,

rather than domain-specific, heuristics. That is, whether one is observing a chemist,

biologist, physicist, or enzymologist, many of the discovery procedures used by the

scientists are the same. Most importantly, Kulkarni and Simon (1988) assert that “the

step-by-step progress of Krebs toward the discovery of the urea cycle” was “produced by a

whole sequence of tentative decisions and their consequent findings, and not by a single

‘flash of insight,’ that is, an unmotivated leap” (174). This conclusion undermines

traditional perspectives towards scientific activity and discovery as “magical” or based on

unobservable insights.

Langley (1981), as well, describes the scientific discovery process as consisting of

general heuristics. The most effective heuristics used by scientists, she writes, lie in their

abilities to detect consistencies and trends in their data, to generate and re-generate

different hypotheses, and to draw on numerous theoretical terms and concepts. And

Bhaskar and Simon (1977), in an earlier study of a chemical engineer solving

thermodynamics problems, characterize scientific problem-solving behavior as a variation

on means-ends analysis, that is, where the scientist observes “a difference between the

present state of the problem and the desired goal state” and works towards reducing the

difference between the two states (203).

While the above characterization of scientists and scientific behavior adapted

from cognitive psychology provides us with many useful insights into the nature of

21

scientific activity, it does, however, de-emphasize two aspects of importance to

researchers interested in the rhetoric of science. First, the literature from cognitive

psychology completely ignores texts and their role in the construction of scientific

knowledge. Second, it de-emphasizes the role that collaboration and group dynamics play

in contemporary science and engineering.

In the next sections, I discuss the literatures from two fields, organizational

behavior and the sociology of science, and outline how they can contribute to what we know

about scientific and technical discourse.

A Description of Science Adapted from Organizational Behavior

As with the literature from cognitive psychology, writing researchers have also

begun to draw on literature from organizational behavior. One interesting line of research

is Harrison’s (1987) work. Harrison argues convincingly that writing researchers and

organizational behavior researchers have numerous interests in common, specifically, an

interest in how context informs a writer’s discourse, beliefs, and methods of argumentation

and consensus-making. She notes that the interaction between cognition and context is a

reciprocal one since writers inevitably shape and alter the organizational environment

surrounding them. A study of the organizational contexts within which writers work,

according to Harrison, should provide researchers interested in writing in the disciplines

with a view of organizations as systems of knowledge and as patterns of discourse which

act and are acted upon by the individuals working in them (cf., Bazerman’s, 1988,

sociopsychological characterization of science discussed in chapter 1). Harrison (1987) also

believes that viewing writing as a situated and complex social activity has important

implications for writing pedagogy. She concludes that “To take seriously the notion of

22

context is to build in students an appreciation for the idiosyncratic nature of terminologies,

semantic systems, beliefs and values, and reasoning processes that characterize any

interdependent social grouping” (19).

Nystrand (1989), too, has called for a definition of composing that views “each act

of writing [as] an episode of interaction, ideally exhibiting intertextuality . . . with a

particular scholarly community or discipline typified by particular premises, issues, and

givens” (70). In his (1989) article, “A Social-Interactive Model of Writing,” Nystrand

elaborates on a view of composing as negotiation. “Communication,” he argues, “begins as

an initial calibration of conversants’ intentions and expectations vis-a-vis the topic and

genre of the text, and the discourse is largely structured by the conversants in terms of each

other’s evolving perspective on the topic and the discourse itself; it is in this sense that we

may speak of discourse as negotiated by the conversants” (73). His perspective towards

writing as situational and social, in turn, foregrounds the importance of contextual

constraints (or what he calls misconstraints) on the writing process. He defines three types

of misconstraints: (a) inadequate elaboration at the level of topic results in abstruse text,

that is, a text that says too much about too few points; (b) inadequate elaboration at the

level of comment results in ambiguous text, or text that says too little about too many

points; and, (c) inadequate elaboration at the level of genre results in misreading (80). This

view towards writing and the contexts within which writing occurs, Piazza (1987) points

out, is based on a conversational model and stresses how “writers must negotiate key text

points and choose appropriate text options since these are guided by the need to share

knowledge and maintain a balance of discourse between reader and writer” (116) (Heath &

Branscombe, 1985; Nystrand & Himley, 1984).

23

Doheny-Farina (1986) also draws on the literature of organizational behavior to

outline his model of collaborative writing. He stresses that our knowledge of how

organizational contexts affect writing can be greatly improved if researchers document in

detail the behaviors of writers in nonacademic and scientific settings. Unlike Harrison

(1987), Doheny-Farina’s (1986) emphasis on the organizational settings within which

writers work, makes him a strong advocate for ethnographic research. To this end, he

argues that

Because any [writing] act can have multiple meanings, researchers seekdiverse interpretations of the acts under study. This can be achieved byexploring participants’ actions from differing points of view, andcollecting data that concerns these actions through several methods(163).

In conclusion, researchers interested in different contexts for writing have begun to

borrow from the field of organizational behavior. Given the social nature of the writing

process, these researchers argue that we must understand group dynamics and their

influence on discourse production.7 And, since an interaction between writing and

organizational structure clearly exists, these researchers advocate a case-study-based

approach to studying writing in context (cf., Doheny-Farina, 1986, 1991), that is, a

methodology that facilitates the collection of detailed information about the

environments which constrain and enable writing in the workplace. As Doheny-Farina

(1986) asserts, such studies

. . . expand the definition of writing activity to include social interactionas a part of the process. Instead of limiting the act of writing to themoments when the writer encodes, research into the social process ofwriting traces the writer’s activities through time, exploring how lifeexperiences impinge upon the rhetorical choices that writer makes (179).

7 See also, Brown and Herndl, 1986, and Herndl, Fennell, and Miller, 1991, for

discussions of communication and miscommunication in nonacademic settings andtheir relationship with organizational structure.

24

A Description of Science Adapted from the Sociology of Science

In addition to drawing on research from cognitive psychology and organizational

behavior, several writing researchers have advocated the need for more “social”

descriptions of writing in academic and nonacademic settings; they argue that the

literature from sociology—particularly the sociology of science and technology—offers

researchers new insights into the behavior of writers in complex organizations. This

literature is so rich, in fact, that Faigley (1985) asserts that “It is tempting to import

wholesale the research issues raised in the sociology of science for the study of

nonacademic writing” (239). Similarly, arguing for the relevance of research from the

sociology of science, Myers (1986) points out that “the sociology of science could provide an

enormous body of evidence for the limitations of the cognitive approach and suggest

methods and perspectives for further work” (596).8 Both Faigley (1985) and Myers (1986),

while acknowledging that sociologists have tended to ignore the role discourse plays in

studies of science and engineering, clearly view research in the sociology of science and

technology as one means for writing researchers to gain a better understanding of the

complex practices of contemporary researchers.

Certainly, not all sociologists of science have ignored the role of the text in

scientific practice and inquiry. Three studies of particular note are Gilbert and Mulkay’s

(1984) “Opening Pandora’s Box: A Sociological Analysis of Scientists’ Discourse,” Law and

8 The debate between cognitivists and social constructivists is an on-going and

complex one, and outside the province of this discussion. Nystrand (1989), forexample, has argued that cognitive theories of the writing process tend toemphasize planning and goal-setting, and ignore the relationship betweenprocesses and products, between reading and writing, and between problem solvingand environmental constraints (see, also, Berkenkotter, Huckin, & Ackerman,1991; Bizzell, 1982a; Bruffee, 1984, 1986; Carter, 1990; cf., Flower, 1989b, foran integrative view towards the cognitive and social dimensions of writing).

25

Williams’ (1982) “Putting Facts Together: A Study of Scientific Persuasion,” and Latour’s

(1988) “A Relativistic Account of Einstein’s Relativity.”

These and other sociologists of science have posited related theories of how

scientists maintain an “objective stance” when composing scientific journal articles. Latour

and Woolgar (1979), for example, describe how scientists must painstakingly add numerous

“modalities”—that is, qualifications—to their texts in order to create personas that are

not claiming to be stating facts. Latour (1988) describes how Einstein’s text can be viewed

using the semiotic constructs of shifting in and shifting out. Scientists attempt, in Latour’s

(1988) opinion, to shift out (i.e., to take the distanced, objective tone of rational observers)

in order to move attention away from them, as actors, and to focus attention on the objects

that they are describing. Finally, Pinch (1985) describes how scientists must negotiate

between internality and externality when constructing their texts; that is, as with Latour

and Woolgar’s (1979) added modalities, Pinch’s (1985) scientists cannot fully externalize

the objects they are describing but, rather, must couch their claims in existing theoretical

and practical knowledge.

The interaction between scientists, their texts, and the scientific claims embodied

by those texts is, however, not an explicit one. Gilbert and Mulkay (1984), for example,

point out that descriptions of the relationship between scientists and their ideas often

ignore the role that texts play in the process. In describing how scientists construct

consensus, Gilbert and Mulkay (1984) posit that scientists attempt to “treat each [scientific]

viewpoint as clearly evident in a scientist’s written products and informal statements” and

to “treat the view of (most) individual scientists as coinciding with one of the current

theoretical labels” (138). In this way, texts—whether they are scientific research articles

or proposals for research funding—are omitted from the picture of the scientific enterprise.

26

Meaning is transferred from the text to the scientist and to his scientific community, and

“theoretical labels” are free of the form by which they are communicated. Texts and

scientists are transformed into theoretical labels and, hence, relegated to the background of

scientific debate.9

Related to this research, sociologists of science who are interested in citation

practices have made similar claims. Gilbert (1977), for example, focusing on the

persuasive nature of citations, asserts that “they provide evidence and argument to

persuade their audience that their work has not been vitiated by error, that appropriate

and adequate techniques and theories have been employed, and that alternative,

contradictory hypotheses have been examined and rejected” (116). In this way, citations

act, not only to justify or account for the positions adopted by the authors, but also to

establish that the authors are forwarding a contribution to the existing literature. Law

and Williams (1982), similarly, describe how scientists “array people, events, findings and

facts in such a way that this array is interpretable by readers as true, useful, good work,

and the rest” (537). For this reason, they characterize scientific research articles as

interpretive networks, as “resources structured in such a way as to induce readers or hearers

to network them in an appropriate manner, ascribing high value to the array itself, low

value to facts that are held to be false, high value to the assumed truth, and so on” (552).

9 See Prelli (1989b) for a compelling exception to this rule. In his case study of the

controversy over the language capabilities of apes, Prelli cites Sebeok’s (1982)systematic attack on Patterson and Linden’s (1981) scientific ethos rather thantheir theoretical arguments. Sebeok asserts that their claims should be ignoredbecause they do not belong to the appropriate scientific community, because theydo not belong to an accredited institution, because they receive “minor grants fromsmall Foundations,” and because they did not “submit their claims forauthorization by competent scientists” (Prelli, 1989b, 56).

27

In conclusion, the literatures from cognitive psychology, organizational behavior,

and the sociology of science offer rhetorical theorists and writing researchers various ways

to begin to describe scientists, scientific settings, and scientific discourse. In the next

section, I provide a brief overview of the historical relationship between rhetoric and

science. In particular, I discuss some of the reasons that science and scientific discourse

have traditionally been outside the scope of rhetorical analysis which, in turn, will

clarify why many of the issues I am interested in exploring are, in part, a result of the

contemporary rhetorical interest in scientific and technical discourse.

The Historical Dichotomy Between Science and Rhetoric

Research in the sociology and rhetoric of science and technology is a recent

development and has posed some difficult and, as yet, unanswered questions. For example,

what are the norms of scientific and technical discourse? How do these norms function? Do

they function in different ways for different scientific discourse communities? How are

academic writers informed by these norms or constraints? How do academic audiences

interpret these norms? One of the major reasons that these are relatively recent questions

is that our contemporary conception of rhetoric differs from traditional conceptions of

rhetoric in two important respects. First, rhetoric is no longer directly associated with

persuasion but, rather, has come to be viewed as a collective, collaborative, epistemic

activity (cf., Burke, 1945). Second, as Burke (1978) has argued, “many questions are called

‘rhetorical’ precisely because there is no ‘truth’ to which one can refer” (16).

Our recognition of the contingent nature of meaning-making has allowed us to

expand traditional rhetorical conceptions of scientific and technical discourse. As Miller

and Selzer (1985) point out,

28

[That] contemporary philosophers and historians of science . . . havecalled into question Aristotle’s separation of the sciences from rhetoricsuggests the possibility of a similarly expanded conception even ascientific one, are often disputed and that syllogistic logic is insufficientto account for the development of a discipline (Kuhn, 1970; Toulmin,1972). Since the premises of science are matters for debate, the discourseof the disciplines (including scientific and technical disciplines) hasbecome a part of the contemporary conception of rhetoric (312).

Tracing this development historically, Zappen (1987) concludes that “Recent

studies in the rhetoric of science and technology owe much of their impetus to research in

the history and philosophy of science and technology, which has undermined the logical

positivist assumption of certainty in science and set in its place notions of science and

technology based upon their concern with probabilities and with audiences, especially

communities of specialists” (288).

Works of particular relevance from the history and philosophy of science include

some of the following: first, Kuhn’s (1970) “The Structure of Scientific Revolutions” is

regarded as seminal because of its assertion that scientific knowledge is not produced

logically and incrementally, but rather, through argumentation, consensus, and

commitment. Second, Toulmin’s (1972) “Human Understanding” explicitly addresses the

decay of positivism and the question of how one chooses a world-view. As Miller and

Selzer (1985) have pointed out, individuals like Kuhn and Toulmin have reinforced the

notion that “Since the premises of science are matters for debate, the discourse of the

disciplines (including scientific and technical disciplines) has become part of the

contemporary conception of rhetoric” (312). And finally, Rorty’s (1979) “Philosophy and

the Mirror of Nature” goes a step further than Kuhn and Toulmin and discusses a l l

knowledge as socially constructed. Knowledge, to Rorty, is “what society lets us say” and

“what . . . is good for us to believe” (12-13).

29

Importantly, a view of science and engineering that is based on the notion of

probabilities and uncertainty, and which stresses the contingent nature of scientific

“truths” and “facts,” does not necessarily result in anarchy or chaos (cf., Feyerabend, 1975);

particularly if that view of science emphasizes the constructive, argumentative nature of

knowledge-making in science and engineering. A rhetorician interested in science and

technology, therefore, is primarily driven to document and understand how research

processes and, more significantly, scientific texts, operate to contribute new knowledge

while simultaneously building community consensus.

Another, related historical development, as numerous researchers have pointed

out, is that traditional definitions of the province for rhetorical inquiry have largely

ignored invention and discovery and focussed, instead, on form and arrangement (Crowley,

1985; Hairston, 1982; Young, 1978, 1987). In Berlin’s (1987) words, “current-traditional

rhetoric . . . was divided into two parts, the first dealing with superficial correctness

(barbarisms, solecisms, and improprieties) and the second with forms of discourse” (37).

And because current-traditional rhetoric emphasized textual elements, it, in turn, ignored

the role of readers, writers, and writing processes. As Berlin and Inkster (1980) assert, “the

tendency of the paradigm [was] to reduce the significance of the writer and to emphasize

the mechanical aspects of composition . . . even when the texts address[ed] a genre that

would characteristically elevate the importance of the writer” (12).

Before turning to the contemporary conception of rhetoric, and its relation to the

study of scientific discourse, it should be noted that these developments in rhetoric are not

entirely the result of developments in the history, sociology, and philosophy of science.

Myers (1986), for example, has correctly pointed out that, until as recently as the 1970s, the

sociology of knowledge concerned itself primarily with “how societies shape . . . systems of

30

belief” and “excluded science from its analysis, assuming that scientific facts were shaped

by method and the natural world, not by society” (597).

Contemporary shifts towards a more “relativistic” view of science and the study of

science (Chubin & Restivo, 1983; Collins, 1981b, 1982; Knorr-Cetina & Mulkay, 1983),

however, have undermined many of the assumptions of the “strong programme” in the

sociology of science (cf., Barnes, 1977; Bloor, 1976). While I favor many of the major tenants

of the new school of sociology, I do so with two important provisos.

First, some contemporary philosophers and sociologists of science run the risk of

over-emphasizing the chaotic, illogical aspects of scientific knowledge-making, thus

promoting a new version of the traditional strong programme’s “black-boxism” (cf., Collins,

1981a; Latour & Woolgar, 1979). That is, similar to composition’s romantic current-

traditionalists, contemporary sociologists of science must be careful not to eliminate

interesting areas of exploration (e.g., invention, creativity, and so on) by describing them as

mysteries without decomposable methodologies (cf., Rothenberg, 1979).10 Second,

although some sociologists of science have argued strongly for the integration and analysis

of scientific discourse (e.g., Gilbert & Mulkay, 1984; Mulkay, 1981; Mulkay & Gilbert,

1982a, 1982b, 1982c; Mulkay, Potter, & Yearley, 1983), the tendency in general has been to

emphasize “unmasking” scientific activity (e.g., Lynch, 1982; Studer & Chubin, 1980;

Travis, 1981), rather than to explore how scientific discourse functions to “mask” the

complexity of that activity. Recently, Collins (1981a) has called attention to the masking

function of scientific discourse when he describes “mechanism[s] of closure,” arguing that

10 See Richard E. Young (1978, 1987) for an excellent overview of the paradigmatic

shift in contemporary rhetoric from the current-traditional approach—whichignores the role of invention and discovery in composing—to new conceptions ofrhetoric which emphasize invention and the social nature of discourse.

31

sociologists need to explore “the use of rhetorical and presentational devices by one group

of experimenters to make their own interpretation of the experimental series the one

credible possibility” (5).

The need for more detailed studies of scientific and technical discourse has been

well-established by numerous researchers over the last fifteen years (e.g., Anson &

Forsberg, 1990; Barnum & Fischer, 1984; Bazerman, 1988; Bernhardt, 1985; Blyler, 1989;

Brown & Herndl, 1986; Debs, 1989; Doheny-Farina, 1986; Fahnestock, 1986; Gilbert &

Mulkay, 1980; Gusfield, 1976; Harmon, 1989; Kelso, 1980; Mayer, 1985; Selzer, 1989;

Weimer, 1977; etc.). In the next section, I describe what researchers interested in scientific

and technical discourse have contributed to our understanding of scientific writers, their

writing processes, and the texts that they produce.

The Contemporary Rhetorical Interest in Scientific Discourse

As I argued in the last section, numerous researchers have begun examining, not only

the individual writer, but also the rhetorical context within which that individual is

situated (e.g., Bartholomae, 1985; Bizzell, 1988; Bruffee, 1986). Most research up to this

point has tended to focus on academic writing as it occurs in the classroom (e.g., Faigley &

Hansen, 1985; Herrington, 1985). However, another group of researchers have begun

examining the nature of writing in the academy, for example, the writing and reading

processes of biologists (Gragson & Selzer, 1990; Halloran & Bradford, 1984; Myers, 1985a,

1985b, 1990), of biochemists (Gilbert & Mulkay, 1984; Rymer, 1988), of engineers

(Herrington, 1985; Selzer, 1983; Winsor, 1989, 1990), and of physicists (Bazerman, 1984,

1988). Indeed, Miller and Selzer (1985), citing Kinneavy (1983), have argued that it is

crucial for rhetoricians to begin to “make some general study of the methodologies,

32

definitions, criteria of evidence, general axiomatic systems, and views of these value

systems” in various disciplines (309). And finally, significant attention has been turned to

writing in nonacademic settings, such as engineering firms, computer companies, hospitals,

and legal institutions (Anderson, Brockmann, & Miller, 1983; Brown & Herndl, 1986;

Harrison, 1987; Moran & Journet, 1985; Odell, Goswami, & Herrington, 1983; Schriver,

1989a; Selzer, 1983, 1989).

Researchers interested in scientific, technical, and nonacademic writing contend

that studying “real-world” writing expands the definition of context by examining the

relationship between writers, their social contexts, and the rhetorical choices available to

them (cf., Doheny-Farina, 1989). Other researchers emphasize how nonacademic writing

helps us to explore the function of discourse communities and how overlapping communities

affect discourse (Odell, 1985). Most importantly, these studies have continued to call for

pedagogical changes that better reflect actual professional writing practices.11

Ultimately, then, studying scientific writing raises a series of interesting questions:

How do academic researchers write proposals for funding and how the the proposal-

writing process inform subsequent research efforts? How does the proposal as genre

empower and constrain academic knowledge and meaning-making? How do academic

researchers represent the audiences who will ultimately grant or refuse them the funding

necessary to continue their research?

11 See, for example, Gilsdorf (1986) who found that business communication

teachers tended to de-emphasize the teaching of persuasive writing despite itswidespread use in business settings, and Barton and Barton (1980) who argue that“The provisional and transactional nature of professional roles must be recognizedby both engineering students and their instructors” (452).

33

These questions anticipate numerous issues currently being explored by

contemporary rhetoricians and writing researchers. Research in professional writing,

scientific and technical discourse, and other non-literary modes of communication, for

example, has stressed the need for studies that describe the relationship between writers’

goals, intentions, plans, and the various texts they produce. This is in response, in part, to

three general developments in contemporary writing research: the emphasis on

(1) examining writing as both a cognitive and a social activity, that is,

studying writing and writers contextually,

(2) describing the process of writing as well as tracing the development of the

written products resulting from those processes,

(3) strengthening the interaction between descriptive studies (where the

intention is to document “what is” current practice in academic and

nonacademic settings) and pedagogical goals (where the intention is to

articulate “what ought to be”).

These three developments, in general, point to the need for what Nystrand (1989)

and Flower (1989) have described as an “Interactive Theory of Writing,” a theory which

integrates the social and the cognitive, products and processes, and descriptive and

pedagogical goals. Writing researchers, Flower (1989) contends, “need what

ethnographers describe as ‘grounded theory’ . . . —a vision that is grounded in specific

knowledge about real people writing in significant personal, social, or political situations”

(283).

34

The Interaction Between Cognition and Context

We now recognize the importance of studying the multiple goals experienced

writers set for themselves based on the complex and changing constraints placed on them in

different writing situations (Hayes & Flower, 1980; Kaufer, Hayes, & Flower, 1986).

What it means to be a “successful” writer clearly varies depending on the writer’s

knowledge base, audience awareness, genre familiarity, and a host of other cognitive and

social factors (Flower, Schriver, Carey, Haas, & Hayes, 1989; Hayes, Flower, Schriver,

Stratman, & Carey, 1987; Selzer, 1989). Broadhead and Freed (1986), too, have argued

that accounting for the interaction between social and cognitive constraints on the writing

process is crucial; unfortunately, they note, “. . . much of the discipline’s focus on cognitive

processes have tended to emphasize the autonomous writer, he or she who composes, not

within a system but by dint of well-oiled heuristics and problem-solving strategies, who

composes by freely negotiating among memory, text, reader, and world” (156). Brandt

(1986), as well, has asserted that the context within which writers compose “remains in

the background of most investigations of writing and writing processes, as a dark stage”

(140).

A notable exception to this is Doheny-Farina’s (1989) case study of an adult writer

working in both academic and nonacademic settings for writing; in the study, he elaborated

on numerous implicit and explicit constraints facing her as she moved from one writing

context to another.12 Certainly Scardamalia and Bereiter (1987) anticipated some of

12 For example, as an academic writer, Anna was expected to act as an authority

figure and to strive for novelty and originality; in the nonacademic community,however, she was faced with a myriad of constraints, such as the external andinternal politics of the organization for which she worked. Anson and Forsberg(1990), similarly, cite “a remarkably consistent pattern of expectation, frustration,and accommodation” on the part of six university seniors as they came tounderstand the differences between academic and nonacademic discourse (200).

35

Doheny-Farina’s (1989) findings when they defined unsuccessful writers as individuals

who produce texts that reveal numerous failings “such as lack of adaptation to audience,

lack of planning, and paucity of revision” (151). Because many of these shortcomings might

well be a factor of the particular setting within which discourse is produced, researchers

have begun to investigate the contextual and social factors that are inevitably brought to

bare on the writing process (Bazerman, 1984; Brandt, 1986; Gilbert & Mulkay, 1980;

Herrington, 1985; Miller, 1980; Odell, 1985; Piazza, 1987).

The Interaction Between Process and Product

The second general development in writing research, as Witte and Cherry (1986)

have pointed out, is that researchers now understand that the process movement in

composition has not been without its shortcomings; they argue that “One of the least

fortunate effects has been the frequent dichotomizing of process and product because of

difficulties encountered in drawing inferences about writing processes from written

products” (114). It appears that, in our eagerness to introduce the process model into

writing research, we have clearly de-emphasized the importance of understanding the

relationship between writing processes, writers, and the texts they produce (Applebee,

1986; Nystrand, 1989; Witte, 1987). Connected to this, Britton (1978) has suggested that “. .

. the most obvious lack [of research on composing] is that of an accurate matching of a fully

revised and edited piece of writing with a complete time record of its production” (28).

There are two possible explanations for why writing researchers have only begun

to emphasize the interaction between writing processes and written products in scientific

and technical settings. First, as I argued earlier, writing researchers have only recently

begun to recognize that such settings are viable areas for rhetorical inquiry. And second,

while there is clearly a need for studies that tie composing processes to the texts they

36

produce (Witte & Cherry, 1986), few writing researchers have drawn on the considerable

research being produced in discourse analysis, text evaluation, and sociolinguistics.13

These studies share a common perspective towards language and language use that

emphasizes finished products and readers’ reactions to those texts. However, because of

this emphasis, researchers often ignore the other end of the communication continuum—the

writers who produce those texts—and instead stress how texts might be improved to

increase reader comprehension. Recently, however, researchers have begun to explore how

our understanding of texts can be tied into our understanding of writing processes (Applebee,

1986; Nystrand, 1989; Witte, 1987; Witte & Cherry, 1986). Finally, having extended our

object of inquiry to account for the entire writer-text-reader continuum, researchers now

recognize the importance of developing methodologies that better capture those

relationships (cf., Doheny-Farina & Odell, 1985; Odell, Goswami, & Herrington, 1985).

The Interaction Between Description and Prescription

The third general development in writing research is, in many ways, a direct result

of our desire to broaden current definitions of composing. Ironically, despite the growing

body of studies describing writing in professional, scientific, and technical settings, some

researchers have complained about how little these studies have influenced writing

pedagogy and practice (e.g., Miller, 1989).

13 While a comprehensive survey of this research is impractical here, some more

notable efforts include Britton and Black’s (1985) research on expository textcomprehension, Charney’s (1987) research on discourse cues and reader-response,Couture’s (1986) collection of essays on functional approaches to writing, Duffyet. al.’s (1982, 1985, 1989) studies of text production and readability, Halliday’s(1978) theory of the ideational functions of texts, Halliday and Hasan’s (1976)analysis of text cohesion, Kintsch and van Dijk’s (1978) model of discoursecomprehension, Schriver’s (1989b) review of reader-oriented approaches to textevaluation, and Vande Kopple’s (1982, 1985, 1986) functional perspective towardsmetadiscourse and the pragmatics of written texts.

37

Again, the split between descriptive studies of writing and classroom practice is

partly historical in nature. That is, the current-traditional paradigm for writing excludes

process theory and, therefore, process studies of workplace composing practices are often

treated as redundant or impractical. Also, as some critics of the process movement have

pointed out, the dichotomy between processes and products continues to be supported in the

writing classroom (Applebee, 1986; Witte, 1987). That is, as advocates for process-oriented

instruction, we are still unclear about how teaching our students about their planning,

drafting, and revising processes actually translates into their finished texts. The major

assumption driving much of our teaching is that giving inexperienced writers knowledge of

“expert” composing strategies will subsequently improve their writing, despite conflicting

research on what exactly it means to “be an expert” (Carter, 1990), and on how much

expertise generalizes across different problem domains (Chi, Glaser, & Rees, 1982; Larkin,

McDermott, Simon, & Simon, 1980) versus how much expertise is particular and situated in

nature (Brown, Collins, & Duguid, 1989; Greeno, 1988; Mehlenbacher, Duffy, & Palmer,

1989; Mehlenbacher, 1992; Suchman, 1987). How, for example, can research on report

writing in different contexts (Brown & Herndl, 1986; Herrington, 1985; Miller & Selzer,

1985; etc.) be used to teach writers from different technical fields? Is there a general

report-writing process or product that transfers across alternative report-writing

situations, for instance, informal reports, reports written for funding agencies, or academic

reports?

In the light of these complex and difficult questions, it is not entirely surprising

that writing instructors have only recently begun to re-examine existing teaching practices.

Students, however, continue to be given style guides, algorithms, and specifications,

despite our awareness that writing is not easily proceduralized. And the emphasis on

38

product-oriented guidelines and writing advice continues to dominate current writing

pedagogy, despite well-documented problems with guidelines and specifications for

writing (Duffy, Mehlenbacher, & Palmer, 1989; Greeno, 1988; LeFevere & Dixon, 1986;

Marshall, Nelson, & Gardiner, 1987).

Alternatives to guideline-based instruction, however, are beginning to get attention

from writing researchers interested in bringing what we have learned about real-world

writing. Spiro, Vispoel, Schmitz, Samarapungavan, and Boerger (1987), for example, have

argued against prescriptive teaching methods in favor of case study-based methodologies.

Further, they point to the need for a theory of case or example sequencing, that is, a

rationale for how instructors order and present model cases to their students.

Yet other researchers have taken a different stance, arguing that students more

effectively transfer knowledge from one problem domain to another if different types of

examples are used as the instruction set (Bassok & Holyoak, 1985). In terms of writing

instruction, therefore, given the goal of producing a proposal for the NIH, the more

example proposals accepted by the NIH that students read, presumably the greater their

chance of producing a successful proposal (at least in theory).

Other researchers have advocated the need for an additional step in the use of

model texts. Mayer (1981, 1985), reviewing two instructional techniques—giving students

concrete models alone versus giving students concrete models and encouraging them to put

the information into their own words—found that groups that put information into their

own words recalled the material better overall.

Unfortunately, many questions regarding the use of model research proposals as a

pedagogical tool remain unanswered. How many models, and of what nature, for example,

39

are optimal for student understanding? Do counter-examples, or “unsuccessful” models,

help writers establish effective principles for future writing contexts? Do some model texts

contain a richer repertoire of principles for good writing and, if so, is there an optimal

number of principles that writers should be exposed to to improve transfer into subsequent

writing contexts? How might researchers who have collected process information integrate

their findings with model proposals?

Because I have collected multiple types of process and product data (see Chapters 4

and 5), I intend to explore how writing researchers—particularly those who have collected

naturalistic data on composing in different contexts—might begin to re-conceptualize

current, “static” guidelines and model texts (i.e., instructional specifications that tend to

ignore different contexts of use and to provide insufficient information on how they should

be applied). What are the pedagogical implications of using case-study data to build

“active exemplars,” that is, examples that integrate process and product information, as

well as pointing to cognitive and social issues that might influence a proposal writer’s

goals and intentions.14 Marshall et. al. (1987), for example, have argued that design

guidelines need to be extended to include examples, counter-examples, and cases where the

guidelines might or might not apply. A detailed case study seems like an excellent

database for writing researchers interested in building effectively “contextual” examples

for general instructional purposes and the teaching of scientific and technical writing.

14 Importantly, I am aware that no single case study of proposal writing is likely to

provide us with a completely representative set of model examples that transferacross different contexts for writing; rather, I view such a case study as a starting-point for documenting how proposal writing in one context differs from and issimilar to proposal writing in other contexts.

40

In the next section, I review some of the existing research on one genre of scientific

writing, the research proposal, and outline a study—the centerpiece of this

dissertation—aimed at contributing to our current understanding of the genre.

Acknowledging the Role of the Proposal in Science

The reasons for the absence of research on proposal writing are varied and complex.

In part, it is because academic discourse has been valued more than other forms of writing

(e.g., proposals, memos, advertising copy, and so on) (cf., Schriver, 1989a). Only recently,

in fact, have researchers begun to acknowledge that proposals may well be “the most

obviously rhetorical writing scientists do” (Myers, 1998b, 1990), and hence deserve further

study.

Yet despite the acknowledged importance of proposal writing in the sciences and

engineering, it is surprising how few researchers have studied the genre (Haselkorn, 1985).

Certainly, studies are beginning to appear (e.g., De Bakey, 1976; Freed, 1987;

Killingsworth, 1983; Mattice, 1984), but few have examined in detail the relationship

between academic writing for publication, scientific research, and proposal writing or, to

use Myers’ (1985b, 1990) expression, “the web of writing” that infuses every scientific and

technical setting.

Moreover, the studies of proposal writing that do exist tend to reinforce many of

the conceptions of scientific and technical writing that writing researchers are currently

trying to undermine. That is, the majority of the research on proposal writing tends to

characterize the activity abstractly and to prescribe guidelines based on anecdotal or

classroom evidence. For example, Mattice (1984) states that “It is important to remember

that while a proposal must be clear and objective, it must also be easily read” (6). Smith

41

(1976) argues that “it should be emphatically stated that the university can do a greatly

improved job of educating [potential proposal writers] in the concepts, terms, and

disciplines of English grammar” (117). De Bakey (1976), in distinguishing between

proposals and written research reports, makes the following assertion:

The temporal relation between the writing and the research, forexample, differs. You write the proposal before you know the results andthe report after. . . . The proposal is a forecast; the final report aretrospection (8).

My contention, from the outset of my investigation of the proposal-writing process,

is that such characterizations of the process reinforce our tendency to devalue the role of

proposals in the construction of scientific knowledge. Certainly they cast proposals as

persuasive, but they disregard the complex interaction between proposals, scientific

research, and scientific journal-article writing. Instead of emphasizing how proposals

influence invention and discovery, they highlight grammatical details. Such

characterizations of proposal and report writing, in addition, reinforce the common

misconception that such writing is factual or objective; as De Bakey puts it, “When you

write the final report, . . . you have the results before you and so can write factually rather

than anticipatorily” (8).

Notably, all the studies that I have cited thus far emphasize written products and

infer process information based on those products. Two other studies, one empirical and one

based on an analysis of forty textbooks, emphasize how readers process and respond to

written proposals. The first study, by Dycus (1977), is generally problematic in that its’

findings contradict much of what we have come to understand about the positive affects of

good document design on reader response (Duffy, Higgins, Mehlenbacher, Cochran,

Wallace, Hill, Haugen, McCaffrey, Burnett, Sloane, & Smith, 1989). That is, Dycus (1977)

42

found that proposal appearance had no significant affect on reader response. He explains

that “After continued reading, the evaluator’s impression of the proposal is based almost

entirely on the thought content of the proposal. At this point, any initial effects of

appearance on impression are almost completely washed out” (291). Improved proposal

appearance, in Dycus’ (1977) words, is “brochure-manship” or “cosmetics.” Numerous

researchers, however, have argued convincingly that attempts to separate the form of a

document from its function are generally problematic (Coe, 1987; Miller, 1984). Dycus’

(1977) stance that the “thought content” of a written proposal can be separated from the

manner within which that content is presented is, therefore, questionable.

The second study, by Freed and Roberts (1989), is a text-based analysis of forty

technical and professional writing textbooks. Disturbingly, their conclusions are that

“little disciplinary agreement exists about what proposals are and how they differ from

some kinds of reports” and that the textbooks “present a bewildering array of classification

systems, often failing to distinguish between situation and function” (317). Finally, they

suggest a view towards proposals, based on schema theory, as event sequences containing

multiple slots for information. This perspective, they assert, should help educators and

writers better understand the proposal genre and function (although they provide no

empirical evidence to support this claim).

The above studies of proposals have several things in common. First, none of them

document the actual proposal-writing process as it occurs in the classroom, laboratory (see

Chapter 3), or workplace (see Chapters 4 and 5). Second, they emphasize the text features

of written proposals out of the context within which they were produced. And third, they

offer pedagogic advice based on anecdotal evidence.

43

At this point, I turn to two studies—one of scientific journal writing and one of

scientific proposal writing—that have significantly informed the direction and design of

the proposal-writing study described in Chapters 4 and 5. The first study, by Jone Rymer

(1988), focuses on how eminent scientists write scientific journal articles. Her selection

criteria for the scientists she studied included “prestigious publication records, high

citation indexes, and extensive [experience] training graduate students” (215). Similar to

my study of proposal writing, Rymer’s scientists defined themselves as “biochemists,

enzymologists, or molecular biologists” (216). Her data are impressive and consist of “17

composing sessions (spanning 3 1/2 months)” as well as notes from seven of the composing

sessions (224).

Among her more notable findings, Rymer states that Scientist-Subject J treated his

“writing [as] a focal point for thinking about the research and making sense of the results,

[for] determining what the science all means” (228). That is, not only does scientific

writing act as a heuristic for discovery, but it also significantly influences the scientific

research being carried out; as Rymer notes,

. . . the planning of the paper is really part of the process ofstructuring the research. . . . The research is being done to reach apurpose. . . . Therefore, the intent to publish a paper is there beforethe research (235).

According to Rymer, her study “shows that scientists are tellers of tales, creative

writers who make meaning and who choose the ways they go about doing so” (244). Of

Subject J’s writing process, she makes the following observations: (1) “Drafting/Revising is

a fused function, the focal point of J’s writing procedures” (232); (2) “Working from the

inside out characterizes J’s drafting process” (233); and, (3) “Overall revision after drafting

is a significant function in J’s process” (235). These findings, it should be noted, conflict

44

with the findings of Broadhead and Freed (1986) and Selzer (1983) who observed that

nonacademic writers rarely revise and tend to compose in a very linear fashion.

Finally, it is useful to contrast my research interests with those of Rymer.

Characterizing one scientist’s composing process, Rymer writes “In carrying out his tasks,

he typifies the behavior of his colleagues (for example, writing longhand on legal pads

[and] submitting drafts for typing section by section)” (221-222); because I was interested in

studying contemporary science “in the trenches” (rather than the writing processes of

“eminent” scientists), my subject was a young, untenured faculty member (see Chapters 4 and

5). And, since my scientist-subject operated in a highly competitive department, it was not

unusual for graduate students and young faculty members to aggressively seek funding for

research and equipment grants.15 The scientist I studied, therefore, did all his composing

on a microcomputer; moreover, his machine was linked to various library and abstract

services. Access to technology that supported his writing and reading processes, not

surprisingly, influenced the nature of those activities. In addition, Rymer explains that

most of her scientists “simply treat[ed] [student] drafts as raw materials to create their own

papers in their own style” (222). In this respect, as well, the scientist I studied controlled

the writing of his proposal from the outset, often incorporating the experimental data

produced by his graduate students and colleagues. Finally, Rymer’s (1988) study focussed

on journal-article writing, the genre which, as I have argued earlier, most rhetoricians and

sociologists of science have privileged (e.g., Bazerman, 1988; Gragson & Selzer, 1990; Gross,

15 As Mukerji (1989) has correctly pointed out, “It is conventional in analyzing the

role of science in modern societies to study well-established physicists orbiologists (i.e., elite members of traditional and prestigious sciences), and usethem as models for the rest of science, even though they are exceptional cases”(14).

45

1985, 1990a, 1990b; Harmon, 1989; Popken, 1988; Swales, 1984; Swales & Najjar, 1987;

Zappen, 1985).

The second (1985b) study, this one focusing on proposal writing in biology, is

contained in Greg Myers’ (1990) compilation of previously published articles and represents

an impressive contribution to the field. Like Bazerman (1988, 1991), Myers’ method of

analysis is based in traditional literary analysis. Citing Culler’s (1968) article, “The

Darwinian Revolution and Literary Form,” Myers points out how significantly the work

influenced his research: “[Culler] identifies texts with the authors as represented in the

text, and imputes to these authorial personae various intentions and interests. In all this,

Culler’s article exemplifies the procedures I will be following in this book” (10). While I

agree with Myers’ motivation to study in detail the texts of scientific and technical

writers, the problem with such analyses is that—in privileging texts rather than the

processes of the scientists who produce them—we can only infer intentionality at best.

Similar to Rymer (1988), Myers (1990) establishes that research, data, and

scientific proposals for research funding are interdependent. Of scientific writing, Myers

contends “a text on a phenomenon takes on the same shape as that phenomenon, or rather,

the phenomenon takes shape through the text” (26). However, while he does discuss the

relationship between proposal writing and scientific research, he does so only generally.

Writing about the composing process of one biologist, for example, Myers (1990) asserts that

“Since he must discuss the alternatives to his model, he becomes more involved with

structure-function relations, if only to dismiss their influence here, so the context of his

research is changed by the process of applying for funding” (56). This interaction, Myers

(1990) contends, is not simply a matter of finding the appropriate expressions to capture the

46

“meaning” of the experiment: “Finding conventional terms for unconventional research is

not just an exercise in rhetoric—it changes the research” (60).

Whereas Rymer focussed on the interaction between drafting and revising, Myers

(1990) emphasizes the types of revising that the biologists engaged in. His conclusions are

that they revised three ways—for “changes . . . improving the readability, defining the

relation to the discipline, and modifying the persona” (47). Although he is not specific

about what form such changes took, his findings regarding revising for an academic

audience are supported by numerous other studies (Atlas, 1979; Bazerman, 1984; Campbell,

1975; Kaufer & Geisler, 1989; Mulkay & Gilbert, 1984).

Finally, both studies are, to my knowledge, the only two thorough analyses of the

writing processes (in the case of Rymer, 1988) and written products (in the case of Myers,

1990) of practicing scientists. Myers’ (1990) points to the gap I am interested in filling

when he writes, “I particularly need data from other disciplines and earlier and later

stages of the publication process, about how persuasion is planned before a draft is written.

. . .” (99). Methodologically, I am influenced by Rymer’s (1988) focus on the writing process

(data collected through protocols and open-ended interviews) and by Myers’ (1990)

emphasis on written products and on close readings of scientific texts. Both studies, I

believe, contribute important information about discourse production in the sciences,

information I am interested in extending with the studies described in Chapters 3, 4, and 5.

General Conclusions

I began this chapter by characterizing scientific proposal writers and reviewing

relevant literature from cognitive psychology, organizational behavior, and the sociology

of science. While these literatures provided us with various, related perspectives towards

47

science and scientific behavior, they all tended to de-emphasize the role of the scientific

research proposal and its relationship with journal-article publishing and scientific

research activities. This allowed me to describe contemporary research aimed as

contributing to our understanding of the following: (1) on viewing writing as both a social

and a cognitive activity; (2) on exploring the relationship between writing processes and

written products; and; (3) on uncovering the relationship between descriptive studies of

writing and prescriptive goals for instruction. Finally, I reviewed existing research on

proposal writing, in particular, focussing on the work of Rymer and Myers.

In the next chapter, I present two pilot studies of proposal writing and research

funding. The first is a talk-aloud protocol study of fifteen professional and technical

writing students producing short research proposals. The second is an interview-based

survey of fifteen academic researchers and emphasizes the contextual aspects of the

research funding process. I conclude by examining the strengths and weaknesses of each

study, and suggest that a long-term, detailed study of an academic proposal writer, in

context (see Chapters 4 and 5), will address many of the shortcomings of the two pilot

studies.

48

Chapter 3—Two Pilot Studies of Scientific Proposal Writing

What he wrote What he meant

It has long been known that . . . I haven’t bothered to look upthe reference.

While it has not been possible to provide The experiment didn’t workout, definite answers to these questions . . . but I figured I could at least geta publication out of it.

Three samples were chosen for detailed The results on the others didn’tstudy . . . make sense and were ignored.

Accidentally strained during mounting . . . Dropped on the floor.

Handled with extreme care throughout the Not dropped on the floor.experiment . . .

Agreement with the predicted curve is:

Excellent FairGood PoorSatisfactory DoubtfulFair ImaginaryCorrect within an order of magnitude . . . Wrong.

It is suggested that . . . it is believed I think.that . . . it appears that . . .

It is generally believed that . . . A couple of other guys think sotoo.

The most reliable results are those He was my graduate student.obtained by Jones . . .

Fascinating work . . . Work by a member of our group.

Of doubtful significance . . . Work by someone else.

Gilbert, G. N. & Mulkay, M. (1984). Opening Pandora’s Box: A SociologicalAnalysis of Scientists’ Discourse. Cambridge, England: Cambridge UP, 176-177.

49

This excerpt introduces an issue of particular relevance to this chapter’s content

and organization. That is, although I undertook both pilot studies with the best of

intentions, I was also aware of several potential shortcomings of both studies from the

outset. For this reason, I conclude the chapter by discussing some of the limitations of each

study in the light of what we do and do not know about scientific proposal writing in

anticipation of my third, and major, study of the genre.

A Talk-Aloud Study of Research Proposal Writing

The Participants

The participants for my exploratory study were 15 technical and professional

writing majors (six senior undergraduate and nine graduate students) enrolled in a

“Computers and Writing” course. All 15 students had written research proposals prior to

taking the course—either as part of another course or as part of an internship. Two out of

the 15 students had proposal writing as part of their job descriptions in previous work

experiences.

The Task

As the first assignment for the course, I asked the students to write a short (2 page)

proposal for a research project designed to inform some aspect of document design. To help

them select an appropriate research question, they were asked to refer to the research

agenda outlined in Schriver’s (1989a) article, “Document Design from 1980 to 1989:

Challenges That Remain” (see Appendix A). The students were told that most effective

proposals include the following generic information: (1) the project’s participants, that is,

50

who will participate in the study, how many will participate, how they will be selected,

and other major demographic characteristics (e.g., age, sex, etc.); (2) the materials used in

the study, that is, laboratory equipment used, its description, etc., and; (3) the methods

used in the study, that is, what you would do and how you would do it (summarize each

step).

Finally, students were instructed to pair off into groups of two and to take protocols

of their partners for 30 to 40 minutes while they wrote their proposals. They were given

instructions on how to do a talk-aloud protocol (see Appendix B). My goals for using the

talk-aloud protocol method were to explore; that is, I was interested in generating a series

of research questions about proposal writing, rather than in confirming any hypotheses

about the nature of proposal writing.

Seven out of the 15 students chose to study the differences between expert and

novice audiences and the role of subject-matter knowledge in reading, probably because

that topic had been the focus of a previous lecture. Six students proposed projects on various

issues such as the difference between oral and written communication, social factors

affecting document design, and the effect of visual design on reading comprehension. Two

students chose related questions on interface design.

Classifying the Protocols

I classified statements in the 15 think-aloud protocols into protocol segments (or

idea units) according to the approach used by Hayes and Flower (1980) and Burtis,

Bereiter, Scardamalia, and Tetroe (1983). Each protocol was then coded according to the

following categories:

51

(1) Perc. Aud .—references to the writer’s perceived audience (i.e., to the

audience that would be reading and responding to the written proposal);

(2) Prot. Aud .—references to the writer’s audience for the protocol (whether

the actual observer or the person who would ultimately analyze the

protocol);

(3) Linear Org .—linear organizational comments (i.e., comments that

indicated the writer was following a linear writing strategy);

(4) Non-Linear Org .—non-linear organizational comments (i.e., comments that

indicated the writer was not following a linear writing strategy); and,

(5) Tone —references to the tone or “sound” of the proposal.

Two of the fifteen protocol transcripts were rated by two raters, to insure

reliability, and then five more transcripts were rated independently, achieving a

reliability rating of 83 percent. The remaining eight protocols were then coded by one of

the two raters.

Since I was also very interested in how the protocol data-collection technique

influenced the task and the writing process, I also coded for three types of “difficulties”

experienced by the participants during their talk-aloud sessions: (a) talk-aloud problems,

that is, references to difficulties experienced talking aloud and thinking or writing; (b)

technical problems, that is, problems with the word processing technology (the writers

were asked to compose using a computer); and, (c) laboratory problems, that is, comments

referring to the artificiality of the task being asked of them.

52

Issues That Emerged from the First Pilot Study

As discussed in the last section, three major interests driving my pilot talk-aloud

study were how professional writers their proposal’s audience, how they organize the

proposal-writing process, and how considerations of tone affect their writing process.

Partic. # IdeaUnits

Perc’dAud-ience

Prot.Aud-ience

LinearOrg.

Non-LinearOrg.

ToneTotal

#Total %

1 178 0 6 7 0 3 16 9%2 306 0 8 10 10 5 33 11%3 175 0 3 5 0 0 8 5%4 102 0 0 13 1 0 14 14%5 152 0 0 22 3 0 25 17%6 146 0 1 5 0 0 6 4%7 138 0 0 8 0 0 8 6%8 173 11 8 5 7 3 34 20%9 113 0 0 2 1 1 4 4%

10 181 0 2 9 0 0 11 4%11 145 0 3 2 7 2 14 10%12 156 5 0 6 0 2 13 8%13 95 0 3 3 0 1 7 8%14 147 0 3 11 0 0 14 10%15 241 0 13 17 0 1 31 13%

Ave 163 1% 3% 10% 2% 1% 16 10%

Table 1: The frequency with which the 15 professional writers considered the audience(perceived versus the audience for the protocol), organization (linearly versus non-linearly), or tone of their proposals.

Table 1 reports the frequencies with which each of these categories occurred based on the

mean scores of two independent raters.16

16 The headings for Table 1 have been reduced to save space. Briefly, they are as

follows: “Partic. #” is the # of the participant and his or her data summary; “IdeaUnits” are the total # of idea units contained in each transcript; “Perc’d Audience”is the # of references to the perceived audience for the proposal; “Prot. Audience”is the # of references to the audience for the protocol transcript; “Linear Org.” isthe # of references to following a linear organization—from participants tomethods to conclusions, and so on—in the transcript; “Non-Linear Org.” is the #of references to following a non-linear organization, i.e., from methods back to theintroduction down to the conclusion, and so on; “Tone” is the # of references to

53

The 15 protocol transcripts contained an average of 163 idea units each, ranging

from 95 to 306. Considerations of audience, tone, and organization took up only 10 percent of

the 15 protocols, while the majority of the protocols consisted of re-reading the written

text and task instructions, sentence level transcription, and sentence level evaluation and

revision. In the next three sections, I review the data in more detail and give examples of

each of the five categories.

Defining an Audience for the Research Proposal

The majority of protocol studies examining the way writers compose have

emphasized planning, transcribing, revising, and evaluation (e.g., Bereiter &

Scardamalia, 1987; Gregg & Steinberg, 1980; Hayes, 1989; Hayes & Flower, 1980; Hayes,

Flower, Schriver, Stratman, & Carey, 1987). Few researchers, however, have attempted

to track how writers construct or represent their perceived audience (cf., Atlas, 1979, Ede &

Lunsford, 1984, Petraglia, Flower, & Higgins, in press, for notable exceptions).

Indeed, the analysis of my 15 protocol transcripts revealed a limited number of

references to the anticipated audience for the proposals; only two of the 15 writers

(participants 8 and 12) directly addressed the audience for their proposal. Ironically, in

addressing their perceived audiences, both students highlighted the fact that they were

producing their proposals out of any meaningful context for production. One student, for

example, defined her audience generally as “the English Department” and then limited it

to the course instructor:

. . . who should I say it’s submitted to, should I say it’s submitted to theEnglish Department or Brad, ummm, is submitted to Carnegie Mellon

the sound or tone of the written proposal; “Total #” is the # of idea units (out ofthe total idea units for the protocol) devoted to all five categories, and; “Total %”is the total percentage of idea units out of the entire protocol transcript.

54

English Department . . . um, Brad, Bradley? No, Mehlenbacher,instructor, computers and writing (Participant 12).

The second student made an explicit reference to the importance of defining one’s audience

and then, given the nature of the proposal assignment that he had been given, created an

“imaginary” audience for his proposal:

All right, okay, I’m writing a proposal, a short proposal for, okay, well,okay, who the hell is this proposal going to? I mean, who am I supposedto be writing for? Am I supposed to be asking for money, or what? Is thisa thesis proposal? I mean, the first thing a writer’s supposed to do isthink about your audience. It doesn’t say a research prop—oh yeah, itdoes. Research project, okay. Uh, I’ll assume it’s for a research projectsuch as the Kellogg project, yeah, the Kellogg project (Participant 8).

His statement is particularly ironic in that his awareness of what it means to be “a

good writer” is at odds with the nature of the task he has been given; that is, the research

proposal task ignores the crucial role that context and audience awareness play in the

composing process.17 This confusion about the general audience for their proposal was

reinforced by the fact that 10 out of the 15 students referred, not to the audience they

believed would be reading their finished proposal, but to the audience of the protocol tape

itself. This tendency to address the audience for the talk-aloud protocol, in general, took

three forms: students either apologized for the way they were carrying out the task, or

interacted with the observer taping them, or censored comments that they felt were

inappropriate for the protocol audience.

17 It might be argued that the lack of reference, on the part of the participants, to the

context or audience for their proposals is an artifact of my particular taskinstructions; i.e., had I stressed a plausible context and audience for the writtenproposal in the instructions, I might have evoked more participant references tosuch issues. I would maintain, however, that any set of task instructions createsthe possibility of emphasizing or de-emphasizing the participants’ problemrepresentations and problem-solving behaviors (cf., Simon & Hayes, 1974;Simon, 1979).

55

Apologies tended to center around how the subjects were performing the writing

task, for example,

This is not what my paper is going to read like. I am just writing as Ithink so I don’t forget anything I said. This isn’t what it’s going to looklike, I promise, because this is really, really ugly (Participant 1).

This is just an outline which is why I am completely grammaticallymixed up here; don’t worry about the way these sentences go (Participant6).

One student revealed an almost painful dedication to carrying out the task properly for the

audience of the protocol:

Mmm, my contact lenses. I have a problem here. I’ve gotta stop. I swearI won’t think about anything while I’m taking out my lenses, but I reallytruly have to take them out. [Tape interrupted]. I’m back, my glasses areon, my eyes don’t hurt anymore, I haven’t thought about anything, andI’ll get back to what I was doing which was beginning to write out aboutmy rough draft on methods (Participant 15).

Interactions with the observer (the person in the room taping the student as he or

she wrote) or with the assumed audience of the protocol were usually either explanatory in

nature (i.e., I am currently performing this action for these reasons, etc.) or conversational

in nature (i.e., am I doing all right? etc.). The following are representative of such cases:

Fine, all right, how we doing on the tape? (Participant 2).

All right, am I doing good? (Participant 8).

For example, comma, ummm, where am I? For example a technicalwriter—oops, spelled technical wrong, backspace to the accidental x—atechnical writer documenting a familiar word-processing packageparenthesis, for example, having used it for six months parenthesis mayfocus on the high-power functions. [To the observer]: It’s probablydriving you crazy when I make mistakes that I don’t notice [laughs](Participant 14).

At this point, it’s twenty after seven. I’ve been working on this sincealmost six-thirty, so it’s about forty-five minutes, which is what youwanted (Participant 15).

56

And finally, their awareness of the observer caused some students to monitor comments

they felt were inappropriate:

Okay, I guess I will just go back and start to re-write this since my typingis so bad, although this is going to be really boring to listen to. Are youlistening anyone? It’s boring (Participant 1).

Okay, umm, the subjects have to be college age, because then that wayyou’ll get, no, they’ll pretty much all have the same ability, well, notnecessarily, some people are stupid. Take that back, I didn’t say that,umm, they’ll be college age, male and female (Participant 11).

Three of the 15 students continually turned to the observer and asked “how am I

doing?” during their writing sessions. This tendency raises questions of observer intrusion,

an issue that clearly has no easy solution. That is, the alternative approach would be to

have the writers take protocols of themselves, which would remove the distracting

observer from the setting, but would introduce the possibility of them falling silent instead

of talking aloud continually.

Approaching and Organizing the Research Proposal

The findings in terms of how the 15 writers approached and organized the writing

of their research proposals were as follows. Nine out of the 15 students organized their

proposal and writing process following the linear formula set out by the task

instructions—introduction, followed by subjects, followed by materials, and so on. The

majority of the students (12 out of the 15) did little or no re-reading until they had

completed an entire section or paragraph of their proposal. Clearly, they viewed their

main goal to be to brainstorm the assignment and, therefore, spent very little time re-

writing or revising, as is the case with the following two students:

57

Okay, I’ll want to come back and, I’ll definitely want to come back andedit this paragraph, but I think I’ll just keep going mainstream, orthought process, so that I’ll get what I’m thinking down (Participant 5).

The purpose, the purpose of this proposal is, I don’t know, I’ll think ofthe exact words later. For this I’ll use an asterisk to denote that I cannotthink of the exact wording, but it is more important to get the ideas downfirst (Participant 8).

This behavior—writing as a brainstorming activity—appeared to be unnatural for some

students. As one student commented,

Okay, ummm, hmm, I feel like I have to re-read part of this before I goon, so let’s take a look at what I have here. Hmm, I am getting a littleantsy because I know I have a lot of editing and re-wording to do, and Iwant to re-organize some of this stuff, but, I’m going to keep typing just toget as much down as I can (Participant 5).

Only three of the 15 students deviated from the assignment’s outline significantly

(i.e., they made as many references to altering the established outline for the proposal as

they did references to following the set outline). In some cases, exceptions to the formula

set out by the task instructions took the form of simply renaming section headings, as in the

following protocol excerpt:

All right, so instead of subjects, I think the first heading I want to haveis . . . how about test groups, test groups, all right (Participant 2).

In another case, the student was obviously more comfortable working on any section of the

proposal she felt required more definition:

What else do we need to write down here? I need to explain my methods,so I need to go back. I’m going back to the beginning of my document andI’m going to go in after I write the question and, before I tell what I’mgoing to do, I’m going to write methods, my method. And there’ssomething else I think I want to call this, procedures section (Participant11).

58

Unfortunately, however, the majority of the protocol transcripts (again, 12 out of

the 15) demonstrated the importance of lower-level text transcription rather than higher-

level organizational concerns. Hence the following student:

Because the level of expertise is subjective, the subjects—that’s reallygood, subjective subjects—will consist of daily, occasional, oops, and newusers to ensure that the re—ooo, misspelled, control i to insert one littleletter—and break, seems like a lot of work—to oopsy, to ensurb toensure—now we have to do that control delete thing—no, then we’llhave to do that control insert—to ensure that the research results are notskewed—great word, skewed—skewed, skewed, skewed, by the use, uh,oh, of a word processing program or hardware platform that may beeither easier or more difficult to learn and use than average comma.Hmmm, I’ve got a lot of fixin’ here. Four different—oh—four differentprograms will be tested on two different platforms (Participant 3).

Certainly, several interesting things are going on in this last excerpt. First, it

reveals the complex interaction between multiple goals for writing. That is, the student is

operating at numerous levels attempting, not only to transcribe the single sentence, but also

to brainstorm his overall research plan. These two goals, in turn, are confounded by

insignificant activities to which he must attend (at least insignificant given his goal of

producing a complete and coherent proposal), for example, word-processor-related

operations.

Constructing a Professional Ethos

Eight of the 15 students clearly recognized the importance of creating a

professional tone or ethos in their proposals. However, because they tended to view the

task as a brainstorming activity, they often drafted using colloquial expressions and made

mental notes to revise later, as the following excerpts exemplify:

When I write this up, a little note to myself, use references like subjectone and subject [laugh] . . . wonderful . . . two (Participant 1).

59

[The author I am referring to] champions the old-fashioned pick-a-number menuing system, in which the user is presented with a simplenumbered list. All right, simple’s kind of prejudicial there, so I’m justgoing to nuke simple (Participant 2).

Two students made explicit references to the importance of “sounding” professional:

Now I hope to devise the reading comprehension test, and that readingcomprehension test should test for . . . speed it should test for. Oh heck, Idon’t know, how about subject satisfaction, and if they get upset theyaren’t going to learn anything, right? And all the other user-friendlythings, and all the other user-friendly criteria, okay, that’s realprofessional [italics added] (Participant 11).

Okay, subjects, okay. I will advertise. . . . Actually, I’m not going to dothis in the first person because it doesn’t sound very professional [italicsadded]. An advertisement will be posted, an advertisement will beplaced in the [school newspaper] asking for volunteers to do anexperiment (Participant 15).

All in all, the students’ references to the tone of their proposals tended to be

relatively superficial in nature. That is, not surprisingly given the assigned nature of the

task, their statements regarding the tone or sound of their proposals were often very

general (e.g., using the “royal we” instead of a personal pronoun, replacing colloquialisms

with formal expressions, and so on). In the next section, I explicate some possible reasons

for the lack of references in the 15 protocols to audience, global organization, and tone or

ethos construction.

Focusing on the “Noise” that Enters Protocol Data

As mentioned earlier, I was also motivated to track how the talk-aloud technique

influenced the 15 writers’ task orientations and writing processes. Certainly, my interest in

examining the “noise” in my data was, in part, political. That is, I was well aware of the

on-going debate between researchers interested in generating hypotheses about composing

60

processes from protocol data and researchers who criticize the protocol methodology. By

looking at where the technique intruded or altered how the 15 students appropriated the

writing task, my intention was to highlight—not only what protocols can do for us—but

also what protocols cannot do for us. To this end, my co-rater and I coded for three types of

recurrent “technical” difficulties:

(a) references to difficulties experienced talking aloud and thinking or

writing;

(b) references to problems experienced with the word processing technology

(the writers were expected to compose using a computer); and,

(c) references to the setting within which the writers were asked to compose.

Table 2 shows the frequency with which such references occurred across the 15 protocol

transcripts:

Participant # ProblemsTalking Aloud

Problems withTechnology

Problems withContext

1 0 0 22 0 2 03 0 44 04 0 2 05 0 0 06 2 0 17 0 5 08 0 2 59 0 1 010 0 4 911 0 0 012 0 2 013 3 0 014 0 1 015 0 0 0

Total 5 63 17

Table 2: The frequency of “technical” difficulties reported by the 15 professional writers.

61

Only two of the 15 students made explicit references to difficulties that they were

having trying to write or think and talk-aloud simultaneously:

It’s hard to talk and type at the same time. It’s really difficult(Participant 6).

[Observer]: What are you thinking?[Participant]: I don’t know. I’m trying to think. And I can’t think out-loud.[Observer]: You’re supposed to think out-loud. It’s a think-aloudprotocol.[Participant]: I’m sorry, it’s my first think-aloud protocol (Participant13).

Nine out of the 15 students had significant technical difficulties including

problems using their computer mice (Participants 7, 9, and 12), difficulties with their

keyboards (Participants 4 and 10), trouble saving their documents (Participants 3 and 5),

difficulties formatting their documents (Participants 5 and 7), and problems with their

computer monitors (Participant 8). Participant 3 had the most significant difficulties,

probably because of her unfamiliarity with the computer software she used to create the

document. Although the interaction between technological medium and the composing

process was not the focus of my research, the intrusion of the word-processing software on

the writing process certainly represents an important factor that requires future

investigation (cf., Duffy, Palmer, & Mehlenbacher, 1992; Hill, Wallace, & Haas, in

press).

Finally, we coded for any references that implied that students found the task

situation artificial or “unreal.” Only four out of the 15 protocol transcripts contained

explicit comments about the nature of the task. Observe the following participant-

observer interaction:

[Participant]: Am I doing the protocol right now?[Observer]: Yes, we’ve been doing the protocol since you started talking.

62

[Participant]: Okay, here I go, I’m doing a protocol. I’m calling up, inWord, a new document. And there it is. I’m sitting in front and I’mstarting timing. We only do this for half an hour. It’s 8:06 and thirtyseconds. Okay, so approximately 9:35, er, yeah, 9:35. . . . All right, hereI go. I’m beginning my assignment. First I gotta figure out what myassignment is. My assignment, I already read. Okay, my proposal isgoing to be on—hey, that little red light blinked [pointing at the tape-recorder].[Observer]: It’s supposed to blink.[Participant]: Why did it, oh, it blinks when I talk. Wheeoooh,wheeoooh. Hey, that’s pretty cool (Participant 8).

One student, after finishing his proposal, described the writing and the proposed research

as “pretend,” and was disappointed because, during the production of the proposal, he had

developed a vested interest in the subject matter:

I’m sorry this is pretend now because I am kind of psyched. I kind of wantto do this. Of course, I could not go to Japan and talk to these people, but Iguess I’d get someone to interpret for me. And that brings in so much more(Participant 1).

One student explicitly described how her “normal” writing activities differed from

that of the protocol session. The most important factor confounding her session turned out to

be her need “to do more research” on the subject she was writing about:

Okay, well, I think I’ve basically covered all of the beginning optionsand now what I would begin to do is I would move off the word-processorat this point and go onto pen and paper, and start to do some research onthis. Since these were my initial thoughts and I came up with thesubjects and the materials and the methods.

[Later].

Okay, one thing, I just went and got the hardcopy and this is the firsttime I’ve ever sat down with the word-processor from scratch and put myideas on the computer. I normally start by pen and paper and doing anoutline and sitting down and doing all the research necessary. It’susually a little more organized than this. This was a little more free-flowing than I usually do it, but it did give me the basic ideas, I think,from where I would proceed (Participant 10).

63

In general, my data suggest that thinking aloud influences the composing process

for at least some writers, although we do not know exactly what that influence means for

researchers studying the process. In addition, it is interesting to note that the technology

used to create the proposals had more of an influence on the writing process of the 15

technical and professional writing students than the think-aloud protocol method.

General Conclusions

To anyone who has watched or listened to writers talking aloud while they

compose, many of these observations may seem at once obvious and secondary to what the

protocol data do tell us about the composing process. My point here, however, is that if

they are so obvious, then it is surprising that so few researchers have addressed the

implications (see Ericsson & Simon, 1980, 1984, and Steinberg, 1986, for notable exceptions).

Most importantly, the above excerpts point to the problematic nature of giving writers

artificial writing tasks. A great many of the negative comments about their writing

experience revealed that they were frustrated by their inability to truly appropriate the

writing task. This frustration was due, in part, to the staged quality of the task itself, to

the lack of a “true” audience for their proposal, and to the lack of a meaningful context for

writing (other than the laboratory context in which they found themselves).

Before turning to the second exploratory study, it is important to return

momentarily to Bazerman’s (1988) description of the proposal-writing process and to the

many questions about proposal writing that, as yet, remain unanswered (see Chapter 1).

While the talk-aloud protocols told us numerous important things about the writing

process—for example, the degree to which issues of audience and tone influence the

students’ writing activities—they told us very little about the various types of negotiation

64

that Bazerman cites as critical. For example, we learned little about the relationship

between scientific and technical research and the proposal-writing process. Because they

involved a laboratory context for writing, and not the actual setting within which

professional writing occurs, we gained little information about the scientific and technical

settings where proposals are generated, read, and evaluated. And, finally, because the

protocols were of single writers producing texts in isolation, we learned little about the role

of collaboration and negotiation that takes place between scientific and technical writers

and their peers, funders, and discourse communities in general.

Bringing Context into the Study of Proposal Writing

The Participants

I conducted the first study to establish some working hypotheses about the

proposal-writing process. A second, interview-based survey was aimed at uncovering

context-specific information about proposal writing and funding in the academy. Its data

consist of a series of open-ended interviews with 15 academic research scientists and

engineers from various disciplines including physics, cognitive psychology, mechanical

engineering, chemical engineering, and computer science.

The Open-ended Interviews

The open-ended interview questions focussed on the process of getting research

funding, perceptions about the differences between funding agencies and the academy, how

academic researchers communicate with research funding agencies and organizations, and

their research backgrounds and interests. Importantly, the second study emphasized the

65

processes surrounding proposal writing and funding activities, as well as the social and

contextual factors that constrain and enable the activity. Again, the study was

exploratory and only aimed at fleshing out some of the disciplinary, institutional, and

cultural constraints confronting academic proposal writers.

In the interviews, I focussed on eliciting information about the research scientist,

his or her research team, facilities and equipment, and perceptions of his or her research

field and agenda. The interviews lasted between one hour and an hour and a half (see

Appendix C for a representative list of the interview questions).

Issues That Emerged from the Second Pilot Study

The Organizational Politics of Proposal Funding

The first and most prevalent representation of the proposal writing and funding

process, by all 15 academic researchers, was that it is a highly political, multi-

institutional enterprise. One researcher, characterizing his personal funding experience,

describes how complex and enormous funding situations can sometimes be:

I think the proposal we had was a five year budget of about 15 milliondollars. We decided what we would do is we would attempt to build afull-scale online library using today’s technology. And . . . they had, Ithink, eight hundred proposals and they funded four and, in fact, itturned out to be pure politics and they didn’t fund any in states wherethere were already engineering centers, and since we had two engineeringcenters in [this state], none of the science centers went into [it], which Ithink was a pity because there were at least two other proposals from

[this university] which were really good.18

18 The term “political,” in this case, is used to describe the affect that a hidden

criterion (i.e., the geographical location deemed appropriate for new centers) had onthe research scientists’ ability to obtain funding from a particular funding agency.

66

The same researcher points out that his being a part of a large project is the result

of years of contacts that span numerous universities and corporations:

. . . when I was at University A, one of our star students introduced me tohis father who was, at that stage, working for ABC publishing company,and his job was involved with electronic printing publishing. Some yearsago, after I was at University B, we wanted to get an online dictionaryfor the computer system. . . . Knowing that ABC publishes dictionaries, Icalled up this fellow . . . and said—to cut a long story short—it’s throughthat contact that we came to the arrangement to put the ABC dictionaryonline. About the time that the project was taking place, what we callthe Uranus Project, he transferred from ABC to GHI corporation, and gotinvolved in trying to set up a group in his area. We knew him already,he was very keen to support our work, so it was a joint contact, which ledto us sending a proposal to GHI.

Also, as one might expect, given the complex and extensive types of negotiation

that are part of the funding process, the academic researchers referred to numerous

situational factors that constrained their ability to successfully obtain funding (e.g.,

yearly budget requirements, where specifically the money is located, etc.):

So it was a lot of people who have been working together in the sameareas, and that’s basically how we did it, and we spoke with our contactand, you know, he visited here, we visited there. . . . He has to put upsome money from his budget and, being new to GHI corporation, he spentsome time trying to figure out what was the appropriate way to go aboutthings. We took his advice and we put in an equipment proposal, whichwould have been last June because . . . they wanted it actually to getfunded from last year’s budget. . . . And they funded it partly from theprevious year and partly from this year. And that went through verysmoothly. . . . He has also funded some specific projects. He’s fundedsome work by two researchers in computational linguistics. That’s not aformal external research grant, that’s from his departmental provisionalbudget.

In the above excerpt, for example, the researcher explains that numerous individuals

collaborated to make the funding situation a reality, that repeat visits were a part of the

process, and that—even though both sides of the negotiation were interested in pursuing a

67

research relationship—proper allocation of funds required intense and creative attention

from the interested funder.

And finally, one researcher talked at length about his experiences with what he

referred to as “experimental funding arrangements.” In one such case, for example, the NSF

refused to fund any proposals that did not provide evidence that the research scientists

could obtain their own, required equipment:

So part of our proposal process was, not only to propose to do the work, butto line up promises of equipment from manufacturers. . . . This was kind ofa new idea for NSF. They were trying to leverage their money in thesense that the head of NSF said, we’re not going to pay for equipmentanymore, we’re going to coerce the manufacturers to give it. So, in manyways, when they first announced this proposal process, when they firstannounced the fact that they were going to give these grants, there werepeople from [four large corporations], they were all there too, even beforewe submitted the proposals. So, ultimately, the way it works is weprepare this proposal and talk with multiple companies about supplyingequipment for it, and they were going to supply the equipment—partlyfor the people doing the software to use, but partly to put this equipmentin critical places in the universities. . . . And ABC corporation, out of allthe companies, . . . was extremely straightforward and supportive in theway they handled this one. . . . And, in this particular case, it wasalmost as if ABC corporation was saying, look, we’re going to support theNational Science Foundation, whoever the universities are who get themoney.

In conclusion, in more than half of the 15 academic researchers’ opinions, getting

proposals accepted had more to do with the political “talk” that took place between

researchers and (interested individuals within) funding agencies than it did with the

actual proposals that researchers eventually submitted for review. Certainly, this finding

confirms what other researchers interested in the funding process have suggested; as

Mukerji (1989) points out

If you look just at the proposals and research reports sent to agencies, itlooks as though the center of the communication system lies in scientistswriting documents for the government about their proposed and actualprojects, but that is not the whole story. Even more information is

68

gathered informally by scientists and agency personnel. Scientists usetheir connections to figure out how to make their proposals as fundable aspossible (62).

This, of course, does not negate the importance of understanding how to produce an

effective proposal but, rather, points to the importance of incorporating oral

communication skills (e.g., the art of negotiation and fact-finding) into the teaching of

proposal writing. That is, proposals are never written in isolation, as students were asked

to do in my first study. Proposals are submitted and funded as part of a complex social

process. The next section discusses the interaction between proposals, funding, and the

research activities of the 15 academic researchers.

The Interaction Between Proposals, Funding, and Research

Proposals are not only intertextual in the sense that they borrow from and are

influenced by a broad corpus of scholarly research; they are also generated and re-

generated across a series of corporate and government funding agencies. As one researcher

pointed out

Yeah, well the [boundaries between different sources of funding are]fuzzy in the sense that, if one project is funding me ten percent, andanother project is funding me ten percent, and another ten percent, andanother one ten percent, I have ideas that transfer across all of them.Okay, so where do I create an idea, you know. If I had this idea . . . thatcould be used in a variety of places, I just list them all [the sources offunding]. I can’t . . . I don’t write down every hour that I’m here, youknow, the three great ideas I had, and this hour was brought to you byABC funding agency and the makers of Stouffer’s frozen food.

Researchers differed on the ethics of receiving multiple sources of funding for their

research. One researcher, for example, felt that how much or little projects overlapped in

69

terms of the research results they produced was an issue that researchers needed to pay

special attention to:

What we would not do is enter into a relationship with a differentfunding agency that focusses on the same kind of problem. We would tendnot to do that. Not that it’s illegal or anything like that, it’s just that . .. sort of two things. One is it wouldn’t make ABC company veryhappy—it’s not proprietary, we’re publishing this stuff—but also itwouldn’t, it’s not clear what the benefit to us would be. It’s sort of easierto work with one motivated partner than to sort of work withcompetitors, each as partners. On the other hand, there are otherfunding agencies that we do talk to and get funding from on very differentthings. Right now we’re talking with some other companies aboutpotential funding for some other work. Quite different. The practicaleffect of that would be very different from the practical effect of this,and so it’s not really in any sense a conflict.

Along with expressing concerns about the audiences for different proposals and

different sorts of research, almost all of the 15 researchers emphasized the proposal-

writing process as a process where achieving group consensus, effectively managing large

documents with multiple authors, and resolving conflicts between the proposal writers and

the intended audience for the proposal played key roles. Perhaps this explains why the

above researcher preferred working with “one motivated partner” rather than with

“competitors, each as partners;” that is, the single relationship allowed him to channel

his communication efforts with a single party and to focus on the research being carried out.

Another researcher characterized his experience writing a large proposal as follows:

Actually what happened, in this case, and this is fairly common, is thatone person will write a draft of the proposal. That is, we all sit aroundin a room and talk about what ought to go in this proposal. Then wemake sort of a one-page outline—what are the key ideas, sections, and soforth. And then one guy goes off and writes a draft of it. And it’s online.And then somebody else will take a turn at going through that draft, andjust change anything they don’t like. Or, if they think that there’s aserious conflict, then they’ll sort of go and argue it out or whatever. Andthen it sort of cycles to the next person. And so it sort of gets iterativelyrefined. And then, usually the person who wrote the first draft is thefirst author on the proposal and really has the strongest statement ofhow the proposal really turns out. Because the rest of the people just sort

70

of push it one way or another. But that’s really standard. The othermode is that you say, you write this section, I’ll write this section, he’llwrite this section, then we’ll glue them together. That happens, but it’sless common.

Although he does not indicate why one collaborative strategy is more common than

the other, the key is that the proposal-writing process described by the above researcher

is one where multiple writers negotiate with each other through a single document. And,

as stated earlier, this conception of proposal writing—as highly collaborative in

nature—has been generally ignored by most contemporary research on proposal writing

(e.g., Budish & Sandhusen, 1989; De Bakey, 1976; Freed, 1987; Freed & Roberts, 1989;

Mattice, 1984). In the next section, I discuss some of the researchers’ observations on an

equally important aspect of proposal writing and funding; that is, the differences they

described between dealing with different funding agencies and organizations.

Differences Between Funding Sources

Five of the researchers discussed their experiences in dealing with two different

types of funding groups. The first observation was that working with corporate funding

agencies and working with government funding agencies represented very distinct types of

interaction. As one researcher stated

. . . with ABC corporation we’re really working very directly, that is,[our contact] is here every two weeks. . . . I mean, he is a participant inthis research project, so that’s very good. The higher bandwidth thecommunication, the better off you are. There’s no doubt about that. Bothin terms of the company being aware of what’s going on, and gettingsomething out of it, and in terms of us understanding . . . the thing is thatwe get something very important out of this relationship that wewouldn’t get if we were funded by NSF or some government agency. Andthat is, we get ABC company’s insight into what are the practicalproblems that they face which, if we could come up with somebreakthrough, would really have an impact. . . . This way, since we’re sotightly coupled, we really get something important out of that.

71

“Having an impact” and solving practical problems was very important to all 15

researchers, and this may be more a function of their disciplines (largely engineering and

computer science) or of the Carnegie Mellon environment. As one researcher pointed out,

Carnegie Mellon is largely a “federation of entrepreneurs.” In this respect, working with

corporate funding groups was described in a much more positive light than might have been

expected. Unfortunately, one researcher lamented, academic scientists are being forced to

turn increasingly to federal and other government agencies for funding:

So, it turns out that, even though maybe ten years ago there was a bigpush to get money from industry, a just sort of movement within thecountry to try to have. In fact, that hasn’t come about, anywhere near asmuch as people were, some people were hoping it would. Including me.

Finally, some of the researchers (four out of the 15) expressed strong concerns that

our North American funding situation is problematic when compared to the habits of

foreign funding arrangements.19 They stressed the need for a process that strengthened the

interaction between them (as academic researchers) and other academic and nonacademic

researchers. One researcher told the following story:

. . . this year, in my research group, we have visiting researchers . . .[from Japanese, British, and French corporations and governmentagencies] come here as visiting researchers. The companies—they’reemployees of these companies—they come at the expense of the companyand spend a year or a year and a half here working on some kind ofresearch project that’s of interest to them, and us, and so what they getout of it is some exposure to a lot of work going on here; what we get out ofit is some very bright people who will become full-time researchers for ayear or more, and who provide very important manpower and brain-power for getting our work done. And it’s a great arrangement. U.S.companies seem to never do this . . . and it just strikes me as a real bigmistake. And I’ve talked about this with people from various U.S.companies, but it’s a very strange situation, because when I tell them

19 Suzanne Roberts, of Applied Technology Associates, Inc., has correctly pointed

out that—despite the U.S.’s huge investment in research and development (morethan $100 billion in 1990)—”. . . the commercial return on our investment isinadequate, particularly with regard to the lion’s share of this tab of about $30billion sponsored by the federal government” (1991, 336).

72

that people are coming here from DEF, and last year we had people fromJKL and various Japanese companies—especially like to do this—thenthe U.S. company people tend to get upset about that. They tend to say,well, you know this is crazy, here’s the federal government and peoplesupporting U.S. universities, and people are coming in from othercountries and getting all the good ideas and then next year there’ll besome product that comes out that markets this idea. Now, at the samethat they’re upset, for some reason, it doesn’t occur to them that theycould do the same thing, the fact that we would welcome them doing thesame thing.

The goal of the second pilot study was not to learn about the entire funding process

but, rather, to find out how academics interact and communicate with funding

organizations. All 15 researchers tended to see the actual research proposal as a very

small part of a much more important and on-going process (e.g., establishing what types of

political and social factors are influencing a funding agency’s interest in funding one branch

of research instead of another, etc.). This was not entirely surprising since the actual

written proposal is a integral part of an extended process designed to support meaningful

research in academic and nonacademic organizations. In the next section, therefore, I will

share some of the researchers’ discussions of the long-term implications of accomplishing

funding arrangements—particularly as it relates to transferring ideas, technology, and

processes into federal and corporate organizations.

The Continuum from Proposal Funding to Technology Transfer

The majority of the interviewed researchers (nine out of the 15) felt that, not only

was it important that communication between academic researchers and potential funding

groups be strengthened, but also that that communication be well-maintained over long

periods of time. The real concern, here, was that ideas and products would “get lost in the

shuffle” if such communication channels were not well-established. In the words of one

73

researcher regarding the minimal interaction between his research group and that of a

funding group:

It worries me, because I would actually like more of an interaction,because I would feel more comfortable about the fact that they arebenefitting from [our research] and, therefore, will continue to support it.And that’s the thing that concerns me. They don’t exercise their optionsto interact with us as much as I’d like to see.

And, surprisingly, the researchers felt such failed attempts to move ideas and products out

of the academic context occurred both with federal and corporate relationships. As one

researcher observed about his interaction in one funding situation:

He [his contract monitor] drops by, it seems to me, two or three times ayear and asks how the research is going. He hasn’t explicitly asked meabout deliverables. This has been a fairly loose arrangement. I thinkhe’s come by with a bunch of people from the group, and they’ve askedme what we were doing, and I gave them an oral report. And they can seewhat we’re doing is approximately what we proposed to do.

Another researcher, describing what she felt had been a very successful funding

relationship, described problems that she had had with other corporations with which

she had worked:

What you quite often find at other companies is you’ll do a project whichthe company sponsors but there’s nobody in the company really interestedin picking up the ideas. So what you do is you do a project which is yourown project and it’s financed by the National Science Foundation orsomething like that, and the company loses the benefit.

The consensus among the researchers, therefore, seemed to be that—while working

with “real” (i.e., nonacademic) problems was very important to them—the difficulty with

interacting with corporate versus government funding groups was that corporate groups

tended not to “think long-term.” One researcher stated this explicitly and described how

some corporations are seeking federal support to overcome this shortcoming:

74

The only difficulty is that, dealing with, only with industry, . . . youhave a difficult time putting together one, coherent program that has along-term focus to it, because all the companies have different views ofhow, what they’d like to see done and, therefore, you can’t put togetherone nice, coherent program, without reasonably trying to get, and ABCcompany as well as all our other sponsors have been helping us to getsupport from the NSF for a long-term push, and then we would have amore long-term push of both the support from ABC company as well asthe National Science Foundation.

Short-term thinking is not caused simply by the general structure of corporate

organizations. Another researcher identified what he felt was an even more major reason

for the problem:

. . . it always boils down to, there’s no one who has on his scheduleanything related to getting ideas from someone else. Nowhere in theirdesign and their development process is there a mechanism fortransferring technology. . . . And I don’t know why that’s true.

Referring to the same corporation, he added

ABC corporation has probably been the poorest at translating what wedo back into the company. And I think that’s because, again, there’s noestablished conduit, there’s no way to get stuff back to them. If, byaccident, someone at ABC corporation happened to discover you’re doingwork and, by accident, they happened to discover it’s relevant to themand, by accident, they happened to get it, that’s fine. And thatoccasionally happened. It never happened to me personally, but I’m sureit will happen to someone. . . . They don’t have in place even themechanisms for tracking what gets done in the universities, and tryingto—you know, I never had a guy from ABC corporation, never in my entirelife, come into my office and say, we would really like to get some of yourideas into our system. It never happened, never ever happened.

Although this section focussed primarily on the outcome of research funding, and

not on the proposal-writing process itself, it did address some of the communicative aspects

of the process that Bazerman (1988) alluded to when he characterized needed proposal-

related research. In the next, and final section, I will discuss some of the issues the open-

ended interviews did not raise.

75

General Conclusions

My second look at proposal writing and the funding process uses the retrospective

accounts of 15 academic researchers. Again, as with the first pilot study, my intention was

not to prove or disprove established hypotheses but, rather, to generate insights into the

proposal-writing process and its role in the larger, institutionalized funding process. This

second survey focussed on the organizational politics of proposal funding, and on the intense

interaction between written proposals, funding processes, and academic research.

Because the second survey relied on open-ended interviews for its data, I was

aware of the constructive nature of the information (cf., Gilbert & Mulkay, 1984; Odell,

Goswami, & Herrington, 1983). As Perkins (1981) has observed, “people really [have]

little access to their mental processes. Instead, people simply report . . . how they thought

they must have done something” (27). Or, as Brown and Canter (1985) warn, “events may

be reconstructed to answer questions that informants may not have conscious awareness of,

or provide an overview of an experience beyond the knowledge of any one participant”

(222). Despite these concerns, the data are still illuminating for several important reasons.

First, the 15 researchers point to the importance of oral communication in the overall

funding process, and they highlight the need for management and organizational skills to

an academic interested in getting research funding. Second, they establish the importance

of viewing proposal writing and research funding as a long-term endeavor, and not as

something that can be studied effectively in isolated writing incidents.

In describing their funding relationships, for example, the academic researchers

rarely limited their remarks to descriptions of the particular proposal documents that had

resulted in their being funded; rather, proposal writing and funding were one part of a

76

greater interaction that involved internal concerns (e.g., departmental- and college-level

as well as research-related) and external concerns (e.g., the number of funding

relationships available to them as well as the need to foster and maintain on-going

relationships). This highlights the difficult challenge that we, as writing researchers,

constantly face—the impossible goal of judging a given document’s effectiveness in the

light of external constraints which may make a well-written document fail or a poorly-

written one succeed (if obtaining funding defines success).

Implications of the Pilot Studies for Research on Proposals

Current developments in writing research have implications for the first pilot

study and for the survey of researchers. That is, contemporary researchers are interested in

expanding our current writer-centered, process-centered view towards writing to account for

the social dimensions of the communication act. Certainly, the debate between researchers

interested in writing as a cognitive activity versus writing as a social act is a well-

established one (see, e.g., Bartholomae, 1985; Berlin, 1988; Bizzell, 1982a, 1982b; Bruffee,

1986; Flower, 1989). However, much of their debate has emphasized theoretical

perspectives and points of contention and ignored the fundamentally different approaches

they use to study the writing process.

For over ten years, for example, cognitive research on writing has benefitted

substantially from protocol studies of writers composing in isolation (e.g., Hayes & Flower,

1980), a methodological technique borrowed form cognitive psychology and computer

science (e.g., Card, Moran, & Newell, 1985; Ericsson & Simon, 1980, 1984; Newell & Simon,

1972). Recently, however, researchers have begun to question the effectiveness of protocol

data (e.g., Cooper & Holzman, 1983; Odell, Goswami, & Herrington, 1983). Cooper and

77

Odell (1976), for example, found that protocols of writers missed important information

such as the writer’s sense of audience and efforts to construct an effective ethos, and

changed depending on whether the task was experimental or “real-world.” My first pilot

protocol study, as well, highlights the need for studies of writers as they produce texts in

natural settings over extended periods of time.

And it is precisely because of this interest in documenting “real-world” settings,

that issues of methodology have re-emerged in the literature. Researchers interested in

studying the writing of academic professionals (e.g., Gragson & Selzer, 1990; Rymer, 1988;

Selzer, 1983; Winsor, 1989) and nonacademic practitioners (e.g., Anderson, Brockmann, &

Miller, 1983; Brown & Herndl, 1986; Harrison, 1987; Moran & Journet, 1985; Odell,

Goswami, & Herrington, 1983), therefore, have supplemented their data-collection

methods with methods borrowed from sociology, anthropology, and ethnomethodology. As

Doheny-Farina and Odell (1985) assert, “If researchers assume that they want to

understand the significance of a given action in a given social context, they will have to do

their research in a naturalistic rather than experimental setting” (506).

Applying this methodological perspective to a study of proposal writing, then,

necessitates that writing researchers enter the laboratories and offices of actual proposal

writers. In addition to collecting talk-aloud protocols of these writers as they compose

proposals for research funding, writing researchers also need to document (through open-

ended, discourse-based interviews, and taped meetings) the departmental and

institutional constraints facing those proposal writers. These techniques, borrowed from

sociology and ethnography, should in turn provide writing researchers with a “thick

description” of proposal writing in a naturalistic setting (cf., Garfinkel, 1967; Geertz, 1973).

Finally, the problem with building a thick description of proposal writing is connected, of

78

course, with time and the density of the context in question—on-going laboratory

activities, journal-article writing, proposal production, evaluation, and implementation

are complex activities and usually take place over years and sometimes decades. Even

conducting a case study of one proposal writer, therefore, has turned out to be an enormous

task (see Chapters 4 and 5).

To summarize, in this chapter I described two pilot studies aimed at helping us

understand the proposal-writing process more fully. Although both studies provided us

with some interesting insights into the general nature of the proposal-writing process, I

concluded by outlining some of the shortcomings of both studies. Finally, I argued that a

naturalistic, long-term case study of scientific proposal writing in context would provide us

with data that both studies de-emphasized.

In the next chapter, I describe in detail the methods used to study a biochemical

engineer as he wrote proposals for research funding. Given the goal of describing proposal

writing as a complex process, while at the same time tying that process to the production of

a finished proposal, my major strategy has been to detail a “complete time record of the

production” (Britton, 1978) of one proposal written and submitted to the NIH. Also,

because a “complete” picture of proposal writing inevitably begins well before the writer

sits down at the terminal and begins writing (and extends well beyond the proposal’s

evaluation and possible acceptance), I collected data on the biochemical engineer’s

perceptions of his research agenda and its relation to the scientific and engineering

community in general, on his research-article writing, and on an unfinished proposal

written six months prior to the proposal under study. Much of my description of the

production of a single research proposal written for submission to the NIH, therefore, is

couched within the broader context of two years of on-going research and writing.

79

Chapter 4—A Case Study from Biochemical Engineering

It is a capital mistake to theorize before one has data.

Sir Arthur Conan Doyle, “Scandal in Bohemia.”

It is assumed that, in the long run, it will be possible to bring together theconclusions of a number of such case studies to formulate a provisional, yetempirically based, social theory of scientific knowledge-production or totest and improve upon existing conjectures about this broad area of socialaction (249).

Potter, J. & Mulkay, M. (1985). Scientists’ Interview Talk: Interviews asa Technique for Revealing Participants’ Interpretative Practices. TheResearch Interview: Uses and Approaches. M. Brenner, J. Brown, & D.Canter (Eds.). NY, NY: Academic P, 247-271.

Then the control aspect. Oh yeah. Can’t have a wonderful techniquewithout controls on it.

First taped meeting of Raymond and Larry, the biochemical engineers.

As outlined in Chapters 1, 2, and 3, the major questions driving my study of how

scientists and engineers write research proposals for funding are:

(1) How do scientists and engineers represent the perceived audience of their

research proposal; that is, how do they characterize the traits of their

intended audience? Do they refer to their audience’s research interests,

sub-specialties, background, biases, values, and so on?

(2) How does the perceived audience for research proposals influence the

organization of the proposal, the content of the proposal, and the

80

presentation of the data reported in the proposal? That is, in the

planning, writing, revising, and evaluation stages of the proposal-writing

process, how frequently do scientists attempt to anticipate potential

audience problems or reactions to their document?

(3) How does the proposal-writing process influence and alter data-collection

techniques and laboratory practices? Does writing a proposal affect the

manner in which scientists present and justify their research activities?

(4) How frequently, and in what particular instances, do scientists incorporate

existing scientific research or literature in their proposals or research

plans?

(5) How frequently do scientists discuss the numerous rhetorical alternatives

or strategies available to them in the proposal-writing process? How do

they adhere to the discourse conventions of their field while at the same

time attempting to represent their laboratory activities in an interesting

and novel way?

(6) How, in general, is the proposal-writing effort managed over time; in

particular, what is the interaction between text and talk in science and

engineering?

These questions were informed, in part, by Bazerman’s (1988) and Myers’ (1990)

discussions of the role of writing in the sciences and, in part, by my experiences collecting

and analyzing the data described in the previous chapter. In particular, both Bazerman

and Myers point to the crucial role that scientists’ representations of the audiences for

81

their texts play in the construction and dissemination of scientific knowledge (see questions

1 and 2). The two pilot studies described in Chapter 3, while providing useful information

about the nature of proposal writing and funding in the academy, de-emphasized the

interaction between written discourse and scientific research activities (question 3), as well

as the inevitable interaction between new scientific texts and the body of existing texts

from which scientist-writers must draw (question 4). And finally, Bazerman and Myers

stressed the importance of examining how scientists come to make the choices they do when

writing and how those choices inform the texts they produce over extended periods of time

(questions 5 and 6).

In this chapter, therefore, I outline the sources of my data and describe my method

of coding and analysis. I begin by describing, briefly, the context within which I collected

my data. Next, I discuss an approach to data collection and design—called Participatory

Design—which influenced my on-going study of a biochemical engineer and his colleague.

Although the data I collected dates back to a writing project undertaken two years ago, the

emphasis of my analysis is on the most recent proposal-writing project (i.e., the three

writing projects I collected data on are presented in reverse chronological order starting

from the most recent and dating back to 1988). Finally, the three writing projects were as

follows: (1) the most recent collaborative proposal-writing project, (2) an earlier proposal-

writing effort, and (3) a journal article written before the two proposal-writing projects

were undertaken.

Data Sources and Collection

The data that I collected to answer the above research questions span two years of

a biochemical engineer’s writing and research career. As with any long-term endeavor to

82

collect detailed information about a complex activity or process, my data represent a

myriad of information types. Most of my data were collected during an intense, three-week

writing period (from September 10th to October 1st, 1990) in which a biochemical engineer

(Raymond) and his colleague from a nearby university (Larry) collaborated on a proposal

for research funding that was ultimately submitted to NIH. The collaborative proposal

project was particularly interesting to me for several reasons:

(1) the biochemical engineer and his colleague were working against a fixed

NIH deadline and, hence, drafted and completed the entire proposal in a

very short period of time;

(2) the biochemical engineer and his colleague felt strongly that the proposal

they were writing represented a significant contribution to the field and an

important extension of their previous research efforts, and;

(3) the biochemical engineer and his colleague felt that the proposal-writing

effort was a particularly challenging one since they were attempting to

merge two, previously unrelated, areas of technology-based research.

The data collected over this three-week writing period are in-depth. Since both

researchers wrote and re-wrote multiple drafts of the proposal on a day-by-day basis, I

was able to collect every draft of the proposal as it evolved from beginning to end (there

were 14 versions in all). In addition, during the three-week period, I was able to collect all

correspondence between the two writers; these were most often in the form of handwritten

slips of paper or electronic mail. Also, I taped two key meetings between the two

collaborators. Both meetings represented important junctures in the writing of the proposal

in that they centered around

83

(1) the methodological implications of the two biochemical engineers’ data-

collection techniques (i.e., the techniques which would provide them with

various types of data and how those data might be organized to form a

cohesive story);

(2) the construction of the objects of study that they would be characterizing

(i.e., 10 model proteins); and,

(3) the management of the overall proposal-writing effort (i.e., who would

write what sections and perform which experiments).

Finally, I interviewed the biochemical engineer for an hour and asked him to characterize

the collaborative effort, in terms of the planning, drafting, revising, and evaluative

aspects of the project, as well as in terms of the problems or difficulties that occurred during

the writing of the proposal.

The collaborative proposal-writing effort was, naturally, significantly influenced

by the biochemical engineer’s on-going proposal- and journal-writing activities. For this

reason, I collected drafts of a proposal that he was writing (and had set aside) when he

began the intense collaborative effort. Along with the drafts of the proposal, I took two

one-hour talk-aloud protocols of the biochemical engineer as he revised what he felt was

an important part of the proposal’s overall argument. In addition, I conducted three

thirty-minute discourse-based interviews with the biochemical engineer which focused on

the proposal’s content and on the “pink slips” (i.e., the reviewers’ comments) received from

an NIH review panel. My interest in his response to the reviewer remarks was predicated

by my assumption that the review process informs subsequent proposal-writing efforts

(Cole, Rubin, & Cole, 1977, 1978; Mitroff & Chubin, 1979; Myers, 1985a, 1985b, 1990).

84

Finally, along with looking at the two proposal-writing projects, I also collected

two drafts of a journal article that the biochemical engineer was composing subsequent to

writing the first proposal. In three open-ended interviews, I obtained the biochemical

engineer’s characterization of the two drafts, his research agenda, and his place in the

general field of biochemical engineering. While the majority of my results will emphasize

the three-week long, collaborative proposal-writing effort, I will draw on data collected

during these earlier journal-writing and proposal-writing efforts.

It should be noted that I believe my methodological approach differs significantly

from “traditional” approaches in the field of writing research in three important ways:

(1) I did not privilege any one data-collection technique. I believe I obtained

reliable information about the proposal-writing process from a number of

sources (i.e., the protocol data provided me with valuable information

about the biochemical engineer’s revision process, the retrospective

accounts provided me with important information about the biochemical

engineer’s sense of the field and of what it means to contribute to existing

research in the field, the numerous drafts I collected provided me with

detailed information about how scientific texts evolve over time, and so

on).

(2) Because I was interested in collecting very detailed information about the

proposal-writing process of a single biochemical engineer, I recognize the

contingent nature of my data; that is, I am careful not to make sweeping

generalizations about proposal writing in science and engineering. My

feeling is that the field of writing research, particularly research on

85

writing as it occurs in naturalistic settings, is still in its early stages and

that we can benefit from collecting numerous long-term case studies such as

this and adding them to a growing database of such studies.20

(3) Since my data collection efforts have been influenced significantly by a

methodological approach called Participatory Design (PD) (explained in

detail in the next section of this chapter), I worked very closely with the

biochemical engineer, incorporating his comments and feedback

extensively, and relying on his skills and knowledge of the field to guide

my analyses.21

In the next five sections of this chapter, I describe the origins of the PD approach

and my incorporation of its working assumptions into my data analysis techniques. I then

describe my methodological approach for coding and analyzing the data collected about

20 In his excellent book, “Writing Biology: Texts in the Social Construction of

Scientific Knowledge” (1990), Greg Myers notes that an administrator who hadread an early draft of the book commented, “What do you expect me to make of astudy with an n of 2?” Myers response was, simply, “. . . with such case studies,what number would be large enough?” (38). My response to the same charge (andmy n is 1 less than Myers’) is a similar one. My major goal in collectinginformation about the writing processes and products of a single biochemicalengineer was to build a “detailed” picture of his composing activities over anextended period of time.

21 The PD philosophy has important implications for the way one approaches thetask of collecting and analyzing data. Most importantly, emphasis is placed onthe conditional, evolving, and interpretive nature of the data analysis task.Because my “study” of the biochemical engineer’s writing process incorporates hisinterpretations of the process, I am less likely to describe behaviors that he,himself, would deny, re-characterize, or flatly disagree with. Incorporating thebiochemical engineer into the research process has been a very positive experience;as Myers (1990) points out, “In contrast to groups of nonscientists to whom I’vespoken, scientists have neither been surprised nor felt threatened by my commentson scientific rhetoric. On the rare occasions when the subjects of these studieshave asked for a change or omission, it was always because I had left room for theimplication that they or someone else was guilty of fabrication, incompetence, orbad management, or where they could be seen as criticizing or mocking otherscientists. Such implications could be seen as unscientific behavior, but mydiscussion of their rhetoric never seems to have been taken as an attack on them asscientists” (248-249).

86

the three interdependent writing efforts under consideration. The first, and most intense,

writing project is the collaborative production of a research proposal written for NIH. The

second project is the biochemical engineer’s previous effort to produce and submit a research

proposal for NIH, as well as his interpretation of the review panel’s comments on the

proposal. The third project focusses on the biochemical engineer’s revisions to a journal

article submitted for publication to an eminent biotechnology journal. These three projects

are described using data from numerous open-ended and discourse-based interviews,

protocols, taped meetings, drafts of on-going writing, handwritten notes, computer

summaries of meetings, and the feedback of the biochemical engineer being studied.

A Participatory Design Approach to Data Collection

One might wonder what a methodological approach, derived from researchers

interested in software engineering and organizational management, has to do with a long-

term case study of proposal writing in biochemical engineering. There are two answers to

this question. The first centers on methodological implications. That is, the PD approach

has evolved as a cooperative effort between two groups that have radically different

goals for working together and that bring to the collaboration significantly different types

of expertise and interests (software developers and users’ organizations). In addition, PD

also aims at finding strategies that can allow continuous interaction between the two

groups (Bjerknes, Ehn, & Kyng, 1987). In the present study, my interaction with the

biochemical engineer has been similarly difficult because the nature of his expertise

differs dramatically from the nature of mine.

The second implication that the PD approach has for a writing researcher

interested in proposal writing is a disciplinary one. That is, an important part of the effort

87

to understand writing as it occurs in the sciences and engineering is the need to familiarize

oneself sufficiently enough to take an “insider’s stance” towards the activities and

discourses that operate in the field under study.22 Just as Floyd (1987) and Thoresen

(1990), two prominent PD researchers, call for a transformation of the basic processes by

which software is developed, I too would argue for a similar transformation of the

processes by which we collect and study data about scientific and technical writing.

Thoresen (1990) criticizes current models of project organization as operating on positivist

assumptions:

Established models for project organization, project work, work analysis,etc., are commonly based on implicit assumptions that the necessaryknowledge somehow exists, making the process of designing systemsmainly a matter of extracting the knowledge from the participants, be itusers or developers. More often than not, the assumptions do not hold.Therefore, development projects need to be transformed from productionprocesses to mutual learning processes. Learning must be built into theprocess, by changing the ways in which project work groups worktogether [italics added] (34).

This perspective clearly emphasizes the iterative and complex nature of design and our

understanding of designs—and, in my case, of collecting data about scientific and technical

writing and of interpreting those data.

A PD approach to the study of scientific proposal writing, therefore, requires that

data be collected and evaluated by both individuals involved in the process—myself and,

in this case, the biochemical engineer. This orientation is not at odds with contemporary

concerns shared by sociologists of science, anthropologists, and ethnomethodologists.23

22 See Lynch (1982) for a pertinent discussion of the implications of taking an

insider’s stance versus taking an ironic stance towards the study of scientists andscientific behavior. Also, see Collins and Harrison (1975) for a related stance.They advocate that sociologists of science need to better educate themselves byactually participating in the experimental and technical process of collecting data.

23 This debate owes much of its historical grounding in the debate between Geertz(1973, 1983) and Clifford (1980, 1982, 1986). See Pearce and Chen’s (1989)

88

Steve Woolgar, in his (1980) article, “Discovery: Logic and Sequence in a Scientific Text,”

for example, begins his textual analysis by stressing the highly interpretive nature of his

discourse analysis. Like Woolgar, ethnomethodologist Michael Lynch (1982) criticizes the

traditional stance taken by sociologists interested in studying scientific activities and

discourse. In his (1982) book, Lynch questions the role of sociologists as impartial

“strangers” who go about observing scientists and categorizing them according to well-

established sociological categories. As with researchers who employ the PD approach,

Lynch and Woolgar view data collection as an evolving and iterative process shared by

both the observer and his or her subjects of study.

Data Coding and Analysis

Given this perspective towards the biochemical engineer and his writing process,

the next four sections describe the data collected on the collaborative proposal-writing

situation, the proposal written prior to the collaborative situation, and the journal article

written by the biochemical engineer before he began writing the two research proposals

(see Table 1 for a brief summary of the data sources and their relation to each of the three

writing projects).

Collaborative NIHProposal

Raymond’s EarlierProposal

Raymond’s JournalArticle

excellent comparison of Geertz and Clifford’s positions regarding the interactionbetween ethnographers and the groups they study. Unlike Geertz, whochampioned “thick description,” that is, “fabricating fictions in order to rendercoherent accounts of exotic cultures” (Pearce & Chen, 1989, 121), Clifford argued“that ethnographic texts involve the on-going ‘give-and-take’ (negotiation ofreality) between the ethnographer and natives instead of the ethnographer’s ownreading and coherent (re)construction of the cultural texts” (Pearce & Chen, 1989,128). For this reason, Geertz described his subjects as “informants,” whereasClifford preferred the term “Indigenous collaborators.”

89

Open-endedinterviews

1 3

Discourse-basedinterviews

3

Notes 12

Protocols 2

Drafts 14 2 2

Taped meetings 2

Table 1: Overview of the various data sources and their relation to each of the threewriting projects.

The method of analysis for these sources of data involved first extracting seventy-

one “episodes” (i.e., segments, individual topics, interview answers, or conversational

excerpts) from the four open-ended interviews, the three discourse-based interviews, the

two talk-aloud protocols, and the two taped meetings. In extracting the episodes for

analysis, I followed Potter and Mulkay’s (1982, 1985) advice regarding the extraction of

passages that focus on specific issues and topics. Extracting passages for detailed critical

analysis is a common practice in social psychology research since, as Mostyn (1985) points

out, “it is not possible for the final report to play back all of the recorded observations”

and, thus, “the researcher must think in terms of condensing, excising, and even

reinterpreting the data, so that it can be written up as a meaningful communication”

(138).24

24 For guidance in extracting and analyzing the 71 episodes from the data that I

collected, I relied heavily on Brenner, Brown, and Canter’s (1985) collection ofessays on qualitative research methods, “The Research Interview: Uses andApproaches.” In that edition, Canter, Brown, and Groat describe how to carry out“restrictive explorations” (81) and Mostyn discusses “culling” one’s data-set tofacilitate interpretive analysis (138-139). I realize that, when extracting episodesfrom a large number of data-types, the major problem to avoid is what George

90

These episodes accounted for 42 percent of the total taped data (based on a word

count), and were first coded for whether they involved planning, revising, or evaluation.

In addition, the episodes centered around two of the three major research issues that I was

interested in exploring (and which were outlined in full at the beginning of this chapter):

(1) audience considerations, that is, how the proposal writer(s) characterized their

perceived audience or how they characterized potential problems or responses that the

intended audience might have with the proposal; and, (2) the relationship between

proposal writing, the scientists’ use of various rhetorical strategies, their understanding of

the discourse conventions of the field, and the on-going scientific research being carried out.

Appendix D gives a detailed breakdown of the numbers and percentages of each episode in

relation to each type of data collected.

Eight out of the 71 episodes, or approximately 11 percent of the episodes

(distributed evenly over the different data types), were coded by two raters, and then

another 27 episodes were rated independently by the same two raters to insure reliability.

Interrater reliability on the 27 episodes (almost 40 percent of the total episodes coded)

reached 86 percent. Forty-seven percent of the episodes were collected during the

collaborative proposal-writing project; 22 percent of the episodes from Raymond’s earlier

proposal-writing effort, and 31 percent of the episodes from his article-writing project.

The majority of the remaining 58 percent of my data emphasized technological

details regarding enzyme purchases, equipment availability, budget allocation plans,

administrative details, and project management.25 For example, although the episodes

(1959) has called “circularity,” that is, defining a series of hypotheses or goals forone’s research and then choosing data that allow one to see what one wants to see(cf., Mostyn, 120-124).

25 Although the pragmatics of scientific experimentation and laboratory practiceswere not the focus of this dissertation, numerous sociologists of science and

91

taken from the open-ended interviews only account for about half of the total data

available, the majority of the unexamined data emphasized the biochemical engineer’s

perceptions of the history of his field and his research background, two subjects that were

not the focus of my study.

In addition, because of my interest in how proposal-writing projects are managed, I

noted any references to project assignments (during the collaborative meetings in

particular) and any references to the general time-line involved in producing a written

proposal (most often made during the open-ended and discourse-based interviews). Since I

had all 14 drafts of the collaborative proposal, I also traced any changes made to the

proposal and contrasted them with the plans made during the two meetings and with the

notes exchanged between the two biochemical engineers.

Analysis of the Collaborative NIH Proposal

As described earlier, the proposal-writing effort that my study emphasizes took

place over a three-week period, from September 10th to October 1st, 1990. Although the

writing of the proposal began on September 10th (when Raymond created a file which

included a brief introduction of the research problem and specific aims, a description of

Larry’s IMAC experiments (Immobilized Metal Affinity Chromatography), and a

description of some of Raymond’s earlier experiments using Differential Scanning

Calorimetry—DSC—techniques), actual discussions between Larry and Raymond about the

idea of writing a collaborative proposal first took place in March of that year. Then, in

the first week of September, Raymond received an announcement from the NIH describing

technology have done interesting research in this area, for example, Collins andHarrison (1975), Knorr-Cetina (1981), Latour and Woolgar (1979), and Woolgar(1981).

92

an upcoming study section on metalloproteins. The Request For Proposals (RFP) stated that

October 1st was the due date for submission, a date which Raymond and Larry were

optimistic about being able to meet.

The data collected during the production of the 22 page proposal consist of 14 drafts

of the proposal which I analyzed in the following ways:

(1) for syntactical evolution, that is, for changes in the average length of

words, sentences, and paragraphs, and for prepositional phrases and

passive voice constructions;26

(2) for organizational evolution, that is, for changes to the text in the form of

additions, deletions, and movements,27 and

(3) for planned evolution, that is, for changes that were the result of specific

interactions either during the two taped collaborative meetings or through

documented notes imbedded in the on-going texts.

The two, taped collaborative meetings lasted approximately two hours each and

clarified why certain collaborative experiments and elements of the proposal were

altered, created, or eliminated. These data, along with the numerous notes and general

comments that the co-authors included in their various drafts, emphasized the “informal”

nature of the collaboration. Along with coding the meetings and open-ended interviews for

audience considerations and so on (described in the above section), I also coded them for

proposal-management details, that is, for any reference by either of the biochemical

26 I used a Macintosh-based program called Grammatik•Mac™ to trace these

alterations across the fourteen drafts.27 I used a Macintosh-based program called OmniProof™ to trace additions, deletions,

and alterations to the fourteen drafts.

93

engineers to planned writing assignments. This allowed me to trace any plans for writing to

the actual drafts produced before and after the two meetings (as well as across the other

drafts).

Analysis of Raymond’s Initial Proposal Effort

To supplement the collaborative proposal-writing data, I interviewed Raymond

about two drafts of a proposal that he was writing prior to beginning the collaborative

proposal effort. The first draft had been created before he received a review from the NIH

panel members, and the second was written following the review. I coded both the

protocols of the writer (taken while he revised a section of the proposal based on the peer

review) and the discourse-based interviews which emphasized different aspects of his

new draft as well as the reviewers’ comments and feedback. As explained earlier,

approximately 40 percent of the episodes I collected from the protocols and the interviews

related to the first proposal were coded with an independent rater to estimate reliability.

Analysis of the Journal Article

The biochemical engineer’s published study of yeast alcohol dehydrogenase

(which he published with four other researchers) is the product of years of research and

writing. For the purposes of my study, I concentrated on two particular drafts—the first

draft that the biochemical engineer sent to the journal for submission and the first re-write,

based, for the most part, on the reviewers’ comments. During three open-ended interviews,

I familiarized myself with the biochemical engineer, his research team, facilities and

equipment, his perceptions of the research field, and his description of the evaluation and

revision processes. The interviews lasted between thirty minutes to one hour each (see

94

Appendix E for a list of some of the questions asked of the biochemical engineer).

Although the questions were general and his answers retrospective, they allowed me to

understand more thoroughly and to evaluate the biochemical engineer’s motivation and

texts (which acted as the focal-point of the interviews). In particular, these meetings

provided me with 20 episodes that contained valuable information about his perceptions of

the biochemical community and his place in it.

Collected Iterations on Data Interpretations

Finally, and in keeping with the PD orientation of my data-collection efforts, I

kept on-going notes of interactions I had with Raymond throughout the process. In

particular, I noted any differences we had in characterizing his writing process. As well, I

gave Raymond two drafts of my results section, and obtained (and integrated) any

comments or feedback he offered. As I pointed out earlier, my incorporation of his feedback

had important methodological and disciplinary implications; my goal was to work with

the biochemical engineer to better understand his writing processes, rather than to attempt

to characterize his writing in isolation. Thus, I encouraged Raymond to freely question,

comment on, or criticize any interpretations I made (verbally or in my written texts) about

his writing processes and products.

In the next chapter, I will discuss the results of the two-year case study. In

particular, the results emphasize how two proposal writers in biochemical engineering

constructed the intended audience for their proposal, how proposal writing influenced

their scientific research, and how frequently they made rhetorical decisions based on their

sense of the potential audience for the proposal and the well-learned discourse conventions

of their field. In addition, I will describe the management of the collaborative proposal-

95

writing project, as well as identifying some of the problems or difficulties that the two

collaborators encountered during that project.

96

Chapter 5—Results of the Case Study

. . . in there I kind of shot myself in the foot. I said, well, it could be animpurity, but I don’t think it’s an impurity, and I just moved on. And . . .the guy that really wanted to torpedo this, highlighted that line. Sothat was an example, you know there’s good things in there, and theexperience, a lot of things we did were defensive in nature. Kind of,yeah, we’re seeing effects, controlled experiment, we’re seeing effects,controlled experiment. But there was one thing that we didn’t even haveto say. It’s an example, if you editorialize too much, the thing gets toolong, it puts a gap between what you see and what you’re saying, and it’s,the best thing is to write down everything in your mind, then go back andrip it out.

Third open-ended interview with Raymond, the biochemical engineer.

What can we know? That is, what can we be sure we know, or sure thatwe know we knew it, if indeed it is at all knowable. Or have we simplyforgotten it and are too embarrassed to say anything? . . . By “knowable,”incidentally, I do not mean that which can be known by perception of thesenses, or that which can be grasped by the mind, but more that whichcan be said to be Known or to possess a Knownness or Knowability, or atleast something you can mention to a friend (28-29).

Allen, W. (1966, 1989). Getting Even. Without Feathers, Getting Even,Side Effects. NY, NY: Quality Paperback.

Raymond’s Background and Research Interests

My major source of data are from an Associate Professor in biochemical engineering.

Raymond did his Ph.D. in biochemical engineering at Cornell University. He is the

recipient of a 1987 Presidential Young Investigator Award, and has been working at

Carnegie Mellon since 1983, where his main research focus has been on the following three

areas:

97

(1) cellular processes, that is, characterizing and developing a protein

expression system for use in high density cell culture; modeling cellular

processes and mapping the properties of metabolic networks;

(2) proteins/enzymes, that is, investigating the molecular-level processes

involved in the deactivation of membrane surface-bound species; and

modulating the transport rate of proteins across membranes through the use

of ligands, and;

(3) bioprocess design, that is, building mathematical frameworks for the

analysis of retrofit problems in bioprocess engineering and integrating cell

biology/biochemistry into process screening and design.

Not surprisingly, the graduate student theses that Raymond is supervising also

reflect his research interests; some of these theses are as follows: “Activity and

Deactivation of Multiple Species of Membrane-Bound Yeast Alcohol Dehydrogenase,”

“Structure-Property Identification in Metabolic Networks,” “Hindered Diffusion of Self-

Associating Proteins,” “a-Chymotrypsin Diffusion in Ligant Gradients,” “Enzyme

Transport in Ligand Gradients,” and so on. Journals that he has published in previously

include “The Journal of Theoretical Biology,” “Biotechnology Bioengineering,” “The

World Biotechnology Report 1984,” “Chemical Engineering Journal,” “Biosystems,” and

“Biotechnology Progress.” He has twenty-two publications in total, dating back to 1984.

With the exception of four articles, all the publications were collaborative efforts.

He has taught courses on advanced heat and mass transfer, unit operations of

chemical engineering, biological transport and pharmacokinetics, and transport

phenomena. In addition, while at Carnegie Mellon, he has developed three new courses:

98

computational biology on network analysis, fermentation technology (which he taught

jointly with a molecular biologist), and biotechnology for technology and people, which is

a first year course for all engineers. Importantly, his major contribution to the technology

and people course has been his emphasis on writing and communication in the sciences and

engineering.28

His continued involvement in the funding process, he points out, reveals—not only

his professional duty as an active academic researcher—but also his on-going desire to “get

a peek at the inside operations of the funding networks.” To this end, he has served as a

review panelist for the Office of Naval Research (June, 1985), for the NSF (September,

1988, and May, 1990), and for the National Research Council (August, 1990).

Indeed, Raymond’s funding activities begin prior to his completing his Ph.D.

These grants take three forms: (1) equipment grants from nonacademic organizations, for

example, Aluminum Company of America, Perkin Elmer Company, and DuPont; (2)

equipment grants from foundations, for example, Keck Foundation, Faculty Development

Grant, and Fisher Life Science Group, and; (3) research grants from government funding

agencies, for example, NIH and NSF. The grants range in dollar value from $5,000 to

$200,000. Not surprisingly, many of his early funding efforts were carried out in conjunction

with his thesis supervisor. Once at Carnegie Mellon, his proposal collaborations began

with senior faculty and have recently shifted to collaborations with other, junior faculty

(Assistant and Associate Professorial level) and his own graduate students (probably a

function of his movement towards seniority).

28 Based on many of our initial conversations, in fact, he designed exploratory studies

to see how his students wrote and read scientific journal articles. His findingsdisappointed him: students wrote and read very linearly, a strategy, he observed,that rarely guided experienced scientific practice.

99

Because my study emphasizes two proposals prepared for a government funding

agency, it is worthwhile listing the progression of government grants that he has received.

His funding began in his first year of his Ph.D. program, May, 1984, when he received a

Research Initiation Grant from NSF to study biochemical reaction networks. NSF funded

him from March, 1986, to August, 1987, to study membrane-bound alcohol dehydrogenase.

From August, 1986, to January, 1988, he and a colleague were funded by NSF to study the use

of ligand gradients for protein separation. He was co-principal investigator with the same

colleague for a large grant from NSF from February, 1988, to July, 1991, to study a similar

technique. And finally, in August, 1989, he received an NIH grant for approximately

$30,000 to investigate NMR and a fluorescence spectroscopy-cultivation system.

Raymond’s Context for Writing

Although I began collecting data on Raymond’s research activities, journal writing,

and proposal-writing goals during May, 1988, the initial proposal-writing effort that my

study describes dates back to December, 1989. This initial proposal, and one of the two

proposals that my dissertation emphasizes, is still in preparation and is expected to be

submitted to the NIH. There are several reasons for Raymond’s delaying to submit the

proposal to the NIH.

First, the proposal, originally entitled “Calorimetric Comparison of

Dehydrogenase Structures,” owes much of its development to an article that Raymond

began writing in May, 1988 (see Appendix F for a chronological summary of the three

writing projects spanning over the two year period). That article, the culmination of

approximately four years of laboratory research and writing, was entitled

“Microcalorimetry and Fluorescence Study of Yeast Alcohol Dehydrogenase: Stability and

100

Heterogeneity Implications,” a title which highlights its relationship with the initial

proposal’s subject-matter. Raymond acted as the article’s main author in collaboration

with four other authors. The journal article was submitted to “Biotechnology Progress”

approximately the same time he began writing the first research proposal.

As Myers (1990) has pointed out, the interaction between journal writing and

proposal writing in science and engineering is often an intense one, as was certainly the case

for Raymond and his colleagues. The article was peer reviewed, accepted pending

revision, and returned to him in July, 1988. However, feeling that the article contained

several problematic gaps, Raymond and his graduate research team turned their attention

back to the laboratory. This re-orientation, in turn, halted his work on the original

research proposal. After four months, numerous experiments, and various drafts of the

journal article, Raymond re-submitted a much revised paper entitled, “Microcalorimetry,

Fluorescence, Fractionation Study of Yeast Alcohol Dehydrogenase: Stability and

Heterogeneity Implications.” The article was sent to the same journal in May, 1989, and

ultimately accepted by “Biotechnology Progress.”

The ultimate title of the journal article reflects an extension in Raymond’s research

approach that he would eventually incorporate into his research proposal; his latest

version of the research proposal and its title—“Comparison of Dehydrogenase Structures

by Microcalorimetry and High Pressure Studies”—in turn, represents one major extension

from the journal article research he was carrying out. The first extension to his research

goals, reflected in the journal article’s title, is an extension of his instrumentation and

data-collection tools, that is, the use of both microcalorimetry and of multiple high-

pressure techniques. His revised proposal, as well, extends the work described in his

101

journal article, and aims to compare numerous classes of dehydrogenase, rather than

studying yeast alcohol dehydrogenase specifically.

While working on the proposal, Raymond was solicited to collaborate with

another group, and that collaboration has now lead to a much-expanded, joint NIH

proposal. The collaborative proposal, as I will refer to it throughout the dissertation, was

completed and submitted to NIH in October, 1990, and is the centerpiece of my case study.

Writing began on September 10th, 1990, and it was submitted for review to NIH on October

1st, 1990. The September 10th version of the proposal (ultimately, I collected all 14 drafts

that the biochemical engineers produced) consisted of a draft of the specific aims of the

research (there were eight aims), the significance of the research (which included a

cursory background and literature survey as well as a draft of the proposed contributions of

the research), the results of preliminary efforts and discussion sections (in terms of the

IMAC experiments and the DSC experiments), the proposed research (the choice of model

proteins and their construction), and two appendices (materials and methods).

The remainder of this chapter is organized as follows. The next section describes

the overall results of the analysis of the 71 episodes (accounting for all three writing

projects). As described in Chapter 4, out of the 71 episodes, 47 percent centered around the

collaborative proposal-writing project, 22 percent centered around Raymond’s earlier

proposal-writing effort, and 31 percent centered around his journal-writing project.

Therefore, after presenting the combined findings of all three writing projects, the next

three sections describe each individual project in detail. Finally, I present the analysis of

the 14 proposal drafts collected from the collaborative proposal-writing project; the

analysis emphasizes, in particular, the interaction between the two biochemical engineers’

plans and goals for the written proposal and the actual text produced over time.

102

Raymond’s “Web of Writing”

In analyzing all three writing projects, together and individually, I characterized

each of the 71 episodes according to the following global categories (also, see Chapter 4):

• the type of data-collection technique used (e.g., taped meeting, open-ended

interview, talk-aloud protocol, etc.), and;

• whether the episodes emphasized the planning, writing/revising, or

evaluating of the document.

Within each of the above categories, I also coded for examples of the following (see

Chapter 4 for reliability ratings):

• Characterizing the audience according to its sub-specialty, background,

values, and so on.

• Anticipating audience reactions to the proposal or to the proposed

research.

• Altering existing research plans to accommodate the writing of the

proposal.

• Integrating existing scientific research (literature) into the proposal or

research plan.

• Identifying technical issues and constraints affecting the data-collection

process.

103

• Discussing rhetorical alternatives or strategies affecting the writing of the

proposal.

The percentage of episodes taken from each of the four data-collection methods

were as follows: 49 percent from the open-ended interviews, seven percent from the

discourse-based interviews, 16 percent from the talk-aloud protocols, and 28 percent from

the taped meetings. My interest in the methodological implications of different types of

data was predicated by Gilbert and Mulkay’s (1984) hypothesis that different methods

solicit different types of responses from scientists (see Table 1 of Appendix G for a

percentage breakdown of the different data-collection techniques in relation to planning,

writing/revising, and evaluating).

I collapsed the writing and revising categories for two reasons. First, the majority

of the data were collected during various types of interview situations (discourse-based

and open-ended), and during the two meetings held by the biochemical engineers. These

forums, of course, involved no actual composing. Second, during the two talk-aloud protocol

sessions (where one would expect original composing to occur), Raymond revised existing

texts and tended to describe his revision process rather than to talk as he transcribed new

text.29

It is important to note, however, that although my data-collection methods did

not elicit more information about the transcription processes of the biochemical engineer or

his colleague, they did provide me with an interesting source of information about how

they planned, revised, and evaluated their scientific texts. The open-ended interviews

29 See Cooper and Odell, 1976, and Cooper and Holzman, 1983, for discussions of

the difficulties that experienced writers frequently encounter in talk-aloudsituations.

104

made up 16 percent of the episodes involving planning (or discussions about planning) and

the taped meetings provided another 28 percent (for a total of 44 percent). Thirty-two

percent of the open-ended interviews and six percent of the discourse-based interviews (a

total of 38 percent) centered on text evaluation (or talk about text evaluation). And, as one

would expect, the talk-aloud protocols provided the most significant source of data for the

writing/revision process (a total of 18 percent). In interpreting these numbers, by the way,

it should be remembered that the overall breakdown into the three writing processes

(planning, writing/revising, and evaluating) is also a function of the percentage of episodes

that made up each of the four data-collection types (see Appendix D for those statistics).

Also, because the open-ended interviews, discourse-based interviews, and taped meetings

involved talk about texts, it is likely that planning and evaluation are more appropriate

venues for these sessions. Thus, the proportion of activities is related directly to the mean

number of words per episode.

In addition to coding for the interaction between the different data-collection

techniques employed and the type of writing process represented by those techniques, I also

coded all three writing projects for how frequently the biochemical engineers’

characterized their audience, anticipated audience reactions, altered existing research

plans, integrated existing scientific research into their texts, identified technical issues,

and discussed rhetorical alternatives. My discussion of Raymond’s writing processes,

therefore, combines detailed descriptions of rhetorical moves that Raymond made when

planning, writing, and evaluating his texts with quantitative summaries of the frequency

with which such activities occurred.30

30 Berelson (1971) has discussed the tension in qualitative research “between

frequencies and real meaning,” a dichotomy which Mostyn (1985) describes as“fallacious . . . since the determination of all categories [for quantitative analysis]involves qualitative judgments in the first instance” (121). Therefore, although

105

Table 2 of Appendix G is a percentage breakdown of the six categories for all 71

episodes. These categories (e.g., characterizing the audience, integrating existing

scientific research, and so on) were also coded for when they occurred during the writing

process, that is, whether they occurred during planning, writing/revising, or the

evaluation of the text being produced.31

Notably, only four of the activities in Table 2 occurred in isolation, that is,

without another activity being simultaneously referred to or carried out. Instances where

the biochemical engineers emphasized one category accounted for the following overall

percentage of the 71 episodes: they characterized their audience in four percent of the

episodes, anticipated audience reactions to their text in four percent of the episodes,

altered existing research plans in seven percent of the episodes, and discussed rhetorical

strategies in 15.5 percent of the episodes. Importantly, discussions of and references to

alternative rhetorical plans and strategies dominated the 71 episodes (71.5 percent) and

played a significant part of all the coded activities (56 percent were co-coded with the

five other categories).

The highest interaction between activities existed between accounting for the

perceived audience of the text and generating rhetorical strategies for producing the text.

That is, audience concerns accounted for 52 percent of the 71 episodes, and the biochemical

engineers appeared to emphasize rhetorical alternatives when accounting for the

audience; to that end, 35 percent of the points made about possible rhetorical strategies

obtaining interrater reliability insures that the defined categories are somewhatgeneralizable, it is important to acknowledge the contingent nature of frequencydata

31 See Flower, Schriver, Carey, Haas, and Hayes, 1989, for a review of the researchon composing and the planning process; Hayes, Flower, Schriver, Stratman, andCarey, 1987, for a review of revision and evaluation, and; Hayes, 1989, and Hayesand Flower, 1980, for a description of the writing process in general.

106

were made in conjunction with points made about the perceived audience of a text. The

following excerpt, from the first taped meeting between the two biochemical engineers,

exemplifies how discussions of audience concerns interacted with discussions of rhetorical

strategies:

Raymond: 1 Yeah. That I think is a good closure, and I think a gooddefensive move too. 2 Because, you know, anybody that spends timereading the proposal, they might react invertibly first, and then they gohome and they sleep on it and come back the next morning, and they getlike, you know, gee-wiz, they didn’t talk about this or that.

Larry: 3 Yeah, I think that’s something we need to talk about at leastand show it, where we’ve considered that, where this metal is.

Raymond: 4 Okay then maybe that could drive the, so maybe we’retalking about the aims list, adding that as sort of an additional aim(Taped meeting).

In sentence 1, Raymond is referring to an argument he and Larry are making in the

written proposal. His reaction to the text is clearly evaluative, and leads naturally into

considerations of audience reaction (sentence 2). Implicit in Raymond’s characterization of

the proposal’s audience is his assumption that they will read and re-read the document

several times looking for potential oversights or shortcomings. In sentence 3, Larry agrees

with Raymond’s assertion and, in sentence 4, Raymond brings the conversation back to the

written proposal, suggesting that they add the argument as an additional aim to the

Specific Aims section of the proposal. In this way, audience orientation prompts the

biochemical engineers’ goal of finding appropriate rhetorical strategies.

Table 3 of Appendix G shows the activities that the biochemical engineers’

engaged in during the planning stage of the writing process. Again, references to

alternative rhetorical strategies dominated the planning episodes (72 percent). Similarly,

the biochemical engineers spent a considerable amount of energy describing and

107

anticipating potential audience reactions to their documents (59 percent). As with the

interaction between audience concerns and rhetorical strategies, audience characterizations

often interacted with audience anticipation:

Interviewer: How did considerations of audience inform the writing ofyour proposal?

Raymond: 1 Yeah it had a bearing on several things because when youlooked at the list, we noticed that they were largely chemists, andmaybe biophysical chemists, and their bent would be very microscopicand molecular. 2 And we spotted the engineers, and the affiliations wesaw, we didn’t detect whether anybody on that group would becollaborating with people we knew. 3 But the message we got wasthere’s going to be chemists there, so we should be pretty careful, in ourdecisions, we’re talking about things that they would probably know alot more about than we would. 4 So either we’d avoid talking about it, orjust kind of expose our ignorance, or do it sloppily, or make sure it gets, itlooks at the right things, and then they would get the idea that we kindof knew what was going on. . . . Based on that it was sort of trying toimagine what other kind of proposals would show up and get reacted to,and where ours was supposed to fit in. 5 Since I saw mostly chemists inthis place, I thought, well, the majority of proposals that were going tobe attractive were probably going to be, probably emphasizing othertechniques and very microscopic stuff using nuclearmagnetic resonance orhigh-powered techniques, a way of looking at molecules, you know,something like that. 6 We were coming at it from a fundamentalviewpoint, but from a more practical overall perspective. 7 And that, Ithink we identified a number of issues that were going to be critical tothe ultimate use of the technology that we were talking about here. 8And, so the bottom line is . . . my thought was we should reallyemphasize two key points throughout the proposal that would probablyset it apart from other proposals that they would be looking at (Open-ended interview).

In the excerpt above, Raymond discusses how the perceived audience for the

proposal influenced the proposal-writing process. Earlier, he had pointed out that the list

of panel members provided by NIH had formed the basis of several discussions between

him and his collaborator. The list, which consisted of the names and institutions of

potential reviewers, did not contain information about their particular specialties,

information that Raymond is quick to supply in sentence 1. Not only does he refer to their

108

research areas, but this reference also informs his understanding of the type of research

they would typically carry out (“very microscopic and molecular”). In sentence 2,

moreover, Raymond reveals that one of his goals for looking at the list of reviewers is a

political one; that is, he points out that none of the reviewers appear to be working with

groups that “would be collaborating with people we knew.” His realization that most of

the reviewers would be chemists also influences the content of the proposal (sentences 3 and

4) and the proposal’s place in the group of other potential proposals submitted to the

reviewers (sentence 5). Finally, Raymond’s characterization of the audience ultimately

informs his sense of how his proposal differs from the other proposals (sentences 6, 7, and

8).

The two key points that Raymond felt would make his proposal stand apart from

the others received by the review panelists were as follows. First, the proposal was about

a separation technique, where the goal was to take an extract containing millions of

proteins from a cell. Their separation technique allowed them to extract a desired protein

out of that cell mixture, which was accomplished by establishing which proteins had

affinities for which metal ions, where the metal ion was mobilized in some kind of

stationary phase. Most proteins that have an affinity for metal ion “absorb back in” or

“pass through” the desired material, so the biochemical engineers could simply target and

“grab” the one that they were interested in analyzing. In addition, they could “tune” the

one that they were interested in through engineering or genetics. Raymond’s belief was

that the reviewers, which were largely chemists, would want to look very specifically at

one protein and one metal ion, and to detail the interaction between the two. He and Larry,

however, in addition to being interested in the principle interaction, were also interested

in background affects; that is, if there were any holes in the background:

109

Raymond: 1 If there’s any holes in the background where you have acontaminated one, it won’t pass through or stick, and you might want toaim the one you’re interested in kind of through that window, kind offishing. 2 So kind of emphasize this window thing, and you’ve got toknow that background; without that background you can’t engineerrationally. And that was one thrust (Open-ended interview).

According to Raymond, the second “thrust” would be much more familiar to the

audience of chemists, that is, characterizing the interaction between the protein and the

metal ion. As Raymond summarizes:

1 Those two things, that’s pretty much what the proposal was hung on. 2There’s a window, use the window, fine-tuning, and then the interactionthing, and try and study that and get a logical set of model proteins. 3Work on more than one protein; work on some that have subtledifferences, and focus on that, and hopefully that model set coversenough situations that you can get some generalities out of it too. 4 Thatwas kind of thinking ahead, but when we saw the audience we kind ofstarted changing the emphasis around a little to try and work this withthat audience (Open-ended interview).

Table 4 of Appendix G represents the activities the biochemical engineers

performed during the writing/revising stage of the composing process. As with the

planning stage, the writing/revising process involved numerous references to rhetorical

strategies (91 percent) and to audience anticipation (17.5 percent). In addition, as one

would expect during this stage, Raymond paid careful attention to how and where to

incorporate appropriate research literature into his text (23.5 percent):

Raymond: 1 I just re-read the section, and it strikes me that the way itwas originally set up was to show how with pressure techniques we canparallel what we propose to do with calorimetry and thereby come toconclusions with two independent methods. 2 But reading it, I think thebest way to change this is to attack the first paragraph, put a couple ofnew sentences in, and then use the examples I already put in here as, justbasically the means to underscore some of the introductory sentences I putin. 3 And then, somewhere in there, try and make a second paragraph,and show how some of the ideas that we’ve discussed, the conclusionswe’ve got from calorimetry, how they didn’t always agree with thisfellow Smith’s studies of these enzymes using the single-pressuretechnique he developed (Talk-aloud protocol).

110

The above excerpt is representative of the many instances in which Raymond

alternated between his knowledge of the field’s scientific research literature and his goals

for the proposal’s line of argumentation. After re-reading a section of the document

(sentence 1), Raymond decided to add some examples to clarify his argument (sentence 2).

As well, he decided to add another paragraph that juxtaposed his calorimetry results

with Smith’s results. Another example of references to the research literature is as

follows:

Raymond: 1 I’ve got a copy of the proposal on one side, and some articleshere, and what I’m going to do is try and re-work section 5.4 a little bit,in light of some articles a referee on a paper provided us with plus inlight of a review article [one of my graduate students] found in thelibrary (Talk-aloud protocol).

This excerpt reveals that writing and reading scientific literature are, for Raymond, very

interdependent activities. Moreover, his understanding of the research of other scientists

is highly strategic; that is, he often defines his own research agenda by setting it apart

from his interpretations of others’ research.32 The following excerpt incorporates

Raymond’s sense of what other researchers are currently doing combined with his “angle”

on where his research fits into that research:

Raymond: 1 There’s been a lot of theory and not too exciting butrepetitive development of kinetic pencil-and-paper models from a groupdown, some school south of here. 2 And some guy’s cranking outtheoretical models and he’s making the assumption it’s there. 3 Andthen there’s some experimental work going on where people showed thatfor some surface-attached molecule preparations that it might be theretoo, although some question marks whether they were interpreting theirdata.

Interviewer: Question marks from you?

32 For an interesting study of innovative behavior in science and engineering, see

Kasperson’s (1976) dissertation. In his study of sixty scientists, Kasperson foundthat innovative scientists tended to pay careful attention to the work of theircolleagues and peers.

111

Yes. 4 Well other people too. 5 It’s very possible but there’s probablyother ways to interpret their data too. 6 But they’re kind of, this is thequestion they’re after so, you know, the results filtered the way they’rethinking about it. 7 So you’ve got guys doing theory, some experimentalwork. 8 We thought that if we could show, at a more fundamental level,the unfolding of a molecule as a complex process or calorimetry couldshow that the denaturation of the whole thing is not a simple one-stepprocess, then there’s some neat evidence here that support the guys doingtheory and add to what these other guys might have done in terms ofexperimentally showing it’s there (Open-ended interview).

Not only does Raymond show a familiarity with the work of other researchers (sentences

1, 2, and 3), but he also points out what he feels are some of the shortcomings of their

research (sentence 3). In sentence 4, he emphasizes that others agree with his assertion

and, in sentences 5 and 6, he explains why he has problems with current research in the

area. In Raymond’s words, the researchers’ experimental question “filtered the way

they’re thinking about it” (sentence 6). Finally, Raymond centers his research technique in

the theory and models of existing research, hence establishing his potential contribution to

the field.

Table 5 of Appendix G shows the activities that the biochemical engineers

engaged in during the evaluation stage of the composing process. Interestingly, almost 20

percent of the evaluation episodes (as with the planning episodes) referred to the need to

re-define existing research plans while writing the proposal:

Raymond: 1 I mean, for the project you have to demonstrate that whathe’s doing in the lab works and it’s feasible and you have to demonstratethat the techniques that we hope to do are feasible and will work. 2 Sobefore that’s done then there’s no possibility for a research plan,actually the way it’s being conducted. 3 And what you’ve got is theproposal and without that there’s no preliminary effort. 4 So I thinkthat they kind of map pretty well. 5 The relative proportions I thinkchange though. 6 Like, for this proposal, since the emphasis was onchrometography techniques, we might have been able to make theproposal a little more hypothetical. 7 But in the proposal it wasprobably more critical for him [Larry] to have decent preliminary datato report and make a decent case for the mutations the protein’s using to

112

pull those off. 8 And we could have talked about some characterizationsstuff in more of a round kind of way, and referred less to the literature forlegitimacy of what we were going to attempt to do in conjunction with hiswork. 9 So with a proposal you kind of need both things. 10 But it’s likethe weighting changes. 11 Before you can actually go ahead, then weneed experiments from my lab to really show that this is a worthwhilething, not just something to pitch in a proposal (Open-ended interview).

Raymond, in characterizing the interaction between the written proposal and the planned

research goals, points out that questions of feasibility drive both activities (sentence 1).

The two activities, therefore, “map pretty well” (sentence 4), with plans being made in the

proposal informing laboratory activities and visa versa. “Hypotheticality,” moreover,

hinges on how much preliminary data they report in their proposal (sentences 6 and 7).

That is, Raymond understands that the more data they are able to report, the less they

have to refer to “the literature for legitimacy” (sentence 8).33 This finding supports the

notion that written proposals are less proposals to “begin” a particular line of research,

and more proposals to extend, refine, or build on research that is currently being carried out

(cf., Bazerman, 1988). And it also undermines the stereotypical notion that proposal

writing occurs “before” the major activities of researching and writing up experimental

findings.

To summarize, Raymond exhibited a broad range of complex strategies in his

writing efforts. In particular, discussions of alternative rhetorical strategies dominated

33 My use of the term, hypotheticality, is consistent with the terms of numerous

other researchers interested in how scientists’ couch or qualify their results in thecontext of existing research frameworks: Harvey (1981) uses the term,“plausibility,” Latour and Woolgar (1979) refer to textual “modalities,” and Latour(1988) discusses how scientists “shift in” and “shift out” in their texts. Gross(1989) has referred to this tension between contributing scientists and the scientificcommunities which surround them as “the paradox of communal competitiveness:while the general progress of scientific knowledge depends heavily on the relativesubordination of individual efforts to communal goals, the career progress ofscientists depends solely on the recognition of their individual efforts” (89).

113

the data and interacted with all five coded rhetorical moves (36 percent of the total

episodes contained an rhetorical move in isolation, while the remaining 64 percent

contained groupings of two rhetorical moves). Table 1 highlights the major interactions

between various rhetorical moves for all 71 episodes:

114

Rhetorical Moves # (% of totalepisodes)

# (% of totalepisodes)

Char. Aud. (1) toRhet. Alts. (6)

4 (40%) 10 (15%)

Ant. Aud. (2) to Rhet.Alts. (6)

11 (61%) 18 (25%)

Alt. Res. (3) to Rhet.Alts. (6)

2 (29%) 7 (10%)

Int. Sci. Res. (4) toRhet. Alts. (6)

6 (75%) 8 (11%)

Disc. Tech. (5) toRhet. Alts. (6)

2 (66%) 3 (4%)

Rhet. Alts. (6) toAnt. Aud. (2)

8 (32%) 25 (35%)

Total 33 (47%) 71 (100%)

Table 1: Overview of the interaction between activities performed by Raymond during allthree writing projects. The table is read as follows: there were 10 episodes (15 percent ofthe total 71) that contained a characterization of the audience (Char. Aud.), and themajority of the characterizations of the audience led to a discussion of availablerhetorical alternatives (Rhet. Alts.)—or 4 episodes out of the 10 (i.e., 40 percent).

As the frequency data indicate, attempts to anticipate the audience most often led

to discussions of alternative rhetorical strategies (25 percent of the total episodes).

Similarly, discussions of rhetorical strategies frequently led to descriptions of audience

expectations and potential problems (35 percent of the total episodes). Importantly,

although it was not in the scope of this particular project, future research will need to

address the quality and effectiveness of these rhetorical strategies in the light of the

reviewers’ and editors’ reactions to the texts being altered.

115

In the next three sections, I describe the results of the analyses of the three writing

projects. The collaborative proposal, created by Raymond and Larry, is the result of an

intense, three-week writing effort, and is described in detail in the next section.

The Collaborative Proposal-Writing Project

Raymond, in characterizing his collaborative proposal-writing project, makes the

following statement:

1 This [collaborative effort] was a bit unique because, in the past, thecollaborations were more that our particular effort was pretty far alongand we needed to get a few experiments done, employing a differenttechnique or something. 2 You’d call somebody up and see if they couldrun a quick experiment, and, by doing that, throw that into what you’vedone and have a more complete package (Open-ended interview).

What made his current experience both challenging and frustrating for him, was the fact

that he was not simply “slotting” the research of another colleague into his current

research agenda or into a well-established line of inquiry; nor was he merely employing

someone else’s laboratory technique or methodology in order to strengthen his research

argument. In this case, he stated

1 We were trying to blend two different universes together and makesomething, you know, a stronger overall package. 2 So there was notreally any working back and forth between two labs. 3 It was a mergingof two independent things (Open-ended interview).

Table 6 of Appendix G gives a breakdown of the various activities performed by

the Raymond and Larry during the writing of the proposal. The collaborative proposal-

writing project consisted of 33 of the total 71 episodes (47 percent). Audience concerns

accounted for almost 50 percent of the episodes and references to the need for alternative

research plans occurred in almost 30 percent of the episodes. The following excerpt

116

exemplifies the interaction between the writing of the proposal and the direction of the

proposed research:

Larry: 1 We did a copper test which was completely lost. 2 I was tellingyou about the quality and so on. 3 By the time we were ready to do theexperiment we found out that it wasn’t set up right. 4 So we’re definitelygoing to have one day.

Raymond: 5 Okay, yeah, shit happens. 6 The discussion of the IMACresults needs work. 7 Then I tried to build a transition between the DSCstuff, so this is sort of a new paragraph I put in. 8 It’s basically based onthe primacy of the IMAC results; however, it was necessary to assess theaffects of the preliminary protein structures. . . . 9 So this is like, we’vegot a protein, drop it in that window, and the IMAC activities studies ofthose proteins would probably show whether the strategy’s working ornot, and close the picture. 10 The thing I’ve been stressing in the DSCsurface localization. 11 And so let’s see if we can perhaps detectlocalization in the DSC, localization binding if possible. 12 To do that,we could do DSC experiments. 13 We’ve seen all this before; it’s prettyhelpful. 14 The only thing I took out was, I took out the isothermal stuffbecause the entropy technique does it anyhow. 15 That’s why I createdthe appendix; extending methods to be employed. 16 Strategies,rationales. 17 So, okay, first let’s choose model proteins and then I triedto do something with the characterization model. 18 Which is DSC kindof shit. 19 Fractionation stuff. 20 Then metal-binding studies (Tapedmeeting).

A discussion of a failed experimental trail (sentences 1, 2, and 3), evolves into a

discussion of which experiments the two biochemical engineers need to carry out and how

they should write them into the proposal. Raymond points out several parts of the

proposal that need to be improved (sentence 6) and describes a transition he has built

between Larry’s IMAC section and his own DSC section (sentences 7 through 11). In sentence

12, he explains the usefulness of the DSC experiments in the context of the proposal’s

argument and, in sentences 14 through 16, he gives his rationale for moving the isothermal

data-collection technique to an appendix. Finally, in sentence 17, Raymond identifies

their need for a definitive list of model proteins.

117

Table 7 of Appendix G shows the percentage breakdown of activities performed by

the biochemical engineers during the planning stage of the proposal-writing effort. The

planning phase of the collaborative project made up 24 of the total 33 episodes (73 percent).

However, the writing/revising phase, as pointed out earlier, was not represented by the 33

episodes. Again, rhetorical considerations dominated the episodes (71.5 percent) as well

as considerations of how the audience might react to the proposal (12.5 percent). One of

the most significant outcomes of the planning sessions between the two collaborators was

their establishing an agreed upon number and type of model proteins and discovering an

effective means of presenting that information in the proposal:

Larry: 1 The reason I’m picking galactosidase is because, from my point ofview, we can say that we already have these [other proteins]. 2 So itwon’t look like we haven’t decided what to do. 3 So, I don’t know, are wegoing to have tails for each of them attached, or. We can do it. 4 We cando five or six different proteins like that. 5 In terms of IMAC that willtake us a while to figure out what’s happening. 6 Especially if we getgood mutations of the cysteine we want. 7 Because we have alreadyestablished that we can do that. 8 We have too many model proteinsnow, that’s the problem. 9 It’s not a problem, but it’s how we want tojustify each . . . variant and for what reason we’re doing it. 10 I can makea case in terms of IMAC. 11 If we show, in terms of IMAC, because it’simportant and also what is the distribution of each of these. 12 In short,I would generate a number of proteins which a different degree to whichwe can qualify those with IMAC. 13 Residual binding to no binding.

Raymond: 14 I guess the problem is, we need a matrix or a table orsomething. 15 That would work a lot for progression, know what I mean?

Larry: 16 Yeah, something like that would put everything inperspective. 17 These are only two cysteines and we are going to makemutations (Taped meeting).

The rationale for choosing to study galactosidase is based primarily on pragmatics;

that is, the biochemical engineers elect to study certain proteins either because they have

access to them (sentence 1) or because the proteins will produce the effects they expect to

use in the proposal’s line of argumentation (sentences 6 and 7). Larry, however, realizes

118

that they were in danger of complicating the proposal by using too many model proteins

(sentence 9) and recommends that they may need to defend their use of several proteins

(sentence 9). He then characterizes how he would defend the model proteins used in his

section of the proposal (sentences 11 to 13). Finally, in sentences 14 and 15, Raymond

recommends that they represent their model proteins using “a matrix or a table or

something,” a suggestion that Larry agrees with in sentences 16 and 17.

Two important issues are revealed by this exchange. First, the proteins that the

biochemical engineers selected for study were chosen, not only because they were of

inherent interest to the scientists, but also because they were readily available. And

second, the decision to represent the model proteins in a table, made during the proposal-

writing effort, would eventually inform the biochemical engineers’ experimental approach

and their eventual journal writing (by providing them with a matrix to basically “fill in”).

Table 8 of Appendix G shows the activities that Raymond and Larry engaged in

during the evaluation stage of the proposal-writing project. The evaluation phase of the

collaborative project consisted of 9 of the total 33 episodes (27 percent). Notably, the

biochemical engineers stressed how the proposal-writing activity influenced their

research activities (33 percent of the total episodes). In the following excerpt, Raymond

describes the difficulties that he and Larry encountered during the writing of the

collaborative proposal:

Raymond: 1 What slowed us down was, we got to the point where we hadthe master document, and I thought my sections were reasonable and hiswere okay, but we got those two things that kind of bogged us down. 2One was relevant to me, the other was relevant to him. 3 And that wasthe definition of a set of model proteins. 4 You know, trying to work thatout as you’re writing. You’re writing. Think about it. 5 And we just had atremendous problem defining a logical set. 6 You know, let’s choose tenmodel proteins. 7 Why we chose them? 8 Why are we working withthem? 9 What do we expect to see with them? 10 What they’re going to

119

give us? 11 And likewise, at that time, I had two possibleinterpretations of some of my results, which didn’t have an impact on theproposal as far as the techniques I wanted to use being useful, but I waskind of fighting with those. 12 I worked out a better interpretation, andthen we got together and made that table there on that blackboard. 13And we talked about it, and suddenly had this idea of the table, andfound a set of things to work with, and went back and wrote our sectionsmore around that thing. 14 Merged them again. 15 But that slowed usdown. 16 That was like, really working on a research plan. 17 Sort oflike, attacked by an inert proposal, so, take a break. 18 Go in, fix thingsup (Open-ended interview).

Raymond emphasizes how important it was for them to establish a set of model

proteins. Moreover, he lists a (well-learned) series of questions that their proposal will

need to answer about their choice of model proteins (sentences 6 through 11): “Why we

chose them? Why are we working with them? What do we expect to see with them?

What they’re going to give us?” In addition, Raymond re-counts the importance of the

table for their proposal’s organization (sentences 12 and 13). And he concludes by stressing

the crucial influence that the proposal had on their research plan: “[Writing the proposal

was] really working on a research plan. [We were] sort of . . . attacked by an inert

proposal” (sentences 16 and 17). Finally, all these issues point to the high interaction

between their representation of the audience for the research proposal and the ultimate

direction that the research plan took; that is, in defining their research plans, the

biochemical engineers could not, if they expected the proposal to succeed, ignore the review

panelist’s opinions, reactions, and criticisms toward their “documented” plan.

In the next section, I outline the analysis of Raymond’s initial proposal-writing

effort which occurred shortly before Raymond and Larry began the intense collaborative

effort described above.

120

Raymond’s Initial Proposal-Writing Effort

The first proposal-writing project consisted of 15 episodes out of the total 71 (21

percent). Interestingly, unlike the collaborative proposal-writing project, Raymond’s

initial proposal-writing effort included no references to audience considerations. It may be

that, when working collaboratively, the biochemical engineers were compelled to

articulate more about the intended audience for their proposal. Or, it may be that,

working in isolation on his proposal reduced the amount of effort that Raymond put into

defining and anticipating his potential audience.

Table 9 of Appendix G represents the activities that Raymond carried out during

the writing of the first proposal. Only one episode in the first proposal-writing project

contained references to Raymond’s planning process. This is most likely a result of the fact

that the data collected about the first proposal project came from talk-aloud protocols and

discourse-based interviews, all which tended to emphasize revision and evaluation. As

pointed out earlier, the talk-aloud protocols were an excellent source of information about

the biochemical engineer’s writing/revising strategies, whereas the discourse-based

interviews tended to elicit evaluative strategies.

Table 10 of Appendix G outlines the analysis of the writing/revising stage for the

initial proposal-writing effort. Eleven episodes out of the 15 (73 percent) were of Raymond

as he revised his proposal. In revising his proposal, Raymond referred frequently to goals

of “making it interesting,” “finding a good angle,” and “covering his butt.” In the following

excerpt, for example, he stresses finding an organization for his argument and building a

convincing justification for his data collection techniques:

121

Raymond: 1 In reading over the other paragraphs in this section, itstrikes me that the logic here has to be turned around here. 2 It’sprobably better to talk about what we do in terms of the pressurespectrum, which is a much different organization than’s here right now.3 So we start moving paragraphs around, and the idea has occurred to methat we’ll be getting some information out of this that might be a bitmore quantitative than we can get from the calorimetry techniques, andthat means I can build in the idea that this is an independent methodgetting towards the answer that we’re after. 4 But also it’ll give me someinformation that I can’t obtain any other way. 5 Which will give us sortof a unique angle and better justification for the use of the high pressuretechniques (Talk-aloud protocol).

Raymond is well aware of the contingent nature of his science. As he states in

sentence 2, he also understands that there are certain advantages to stressing one particular

data-collection approach (pressure spectrum) over another (calorimetry techniques)

because the former will provide him with more quantitative results (sentence 3). He

concludes by pointing out that this revision will strengthen his argument by doubling the

number of independent methods used to observe the phenomenon (sentence 5). Indeed,

Raymond’s revision processes are often very strategic and rhetorical:

Raymond: 1 So in revising this section here, it’s probably about the samelength as the section was originally, but it now contains a little morebackground information, which demonstrates that we know what we’retalking about and what we hope we can get done. 2 And it accomplishesthis [reference to the need for a] facility at the University of XYZ. 3 Andthe only thing that we might need in the future, or this proposal mightneed more of is some mathematics and some theoretical frameworks tothrow in, which would probably make it a little harder to read, but itmight be useful to put some of those things in, in any case, just to showthat we can use the theoretical frameworks out there as data analysistools (Talk-aloud protocol).

Sentence 1 reveals Raymond’s well-learned knowledge of scientific discourse

conventions; that is, he is aware of the importance of providing the necessary background

information in order to frame his potential contribution to the field. In sentence 2, he refers

to another rhetorical strategy, the reference to his wanting access to facilities located at

122

the University of XYZ. As Raymond posits in another talk-aloud protocol episode, “I just

mention the keyword that this laboratory is an NIH resource and that it’s there for people

like me to go out and fool around at, and hopefully that will be a positive factor in the

reaction these people have to this section of the proposal.” Raymond concludes the excerpt

by suggesting that the proposal may ultimately need “some mathematics and theoretical

frameworks” which, despite making the document more difficult to read, should heighten

the proposal writer’s credibility.

In conclusion, although only four out of the 15 episodes (27 percent) involved text

evaluation, these episodes often revealed Raymond’s awareness of how proposals are

reviewed and evaluated, particularly in collaborative situations versus individual efforts.

Most importantly, Raymond states that reviewers often look for the proposal writers’

levels of commitment to other projects:

Raymond: 1 When you review a paper that’s collaborative, you tend notto really notice the collaboration. 2 When you review a proposal that’sa collaborative effort, you tend to take more note of it; that’s because youwant to ascertain, in your own mind, whether the collaboration’smeaningful and what are the odds of it actually being able to take place.3 Yeah, there’s a real reinforcement going on. 4 People do look at that. 5Especially really big proposals. 6 Especially NIH, where proposals tendto be much bigger, more money and personnel than NSF, and people will,on the panel, look at that in the end, after they talk about the proposal.7 You know, comments like, why’s so and so written into this thing; hedoesn’t seem to be doing much, and that would take 200k off the budget orsomething. 8 So, and the comments that the reviewer would get backwould be, not necessarily those comments, they might be an overallreduction in the budget and kind of leave it up to the PIs [PrincipalInvestigators] to decide who gets axed (Open-ended interview).

Raymond views collaboration in the proposal-writing process as distinct from

collaboration in journal-article writing (sentence 1); in addition, the size of the proposal’s

budget (sentence 5) and the funding agency being applied to (sentence 6) influence how

carefully the reviewers will attend to the collaboration. Finally, he emphasizes the

123

contingent nature of a proposal’s success, stating that restricting the budget is sometimes one

way of limiting a proposed collaboration (sentences 7 and 8).

In the next section, I outline the results of the analysis of the journal-writing

project. In particular, I was interested in identifying any notable differences between

composing a scientific journal article and writing a proposal for research funding.

Raymond’s Earlier Journal-Article Project

Twenty-three of the 71 episodes (32 percent) made up Raymond’s journal article

project. All 23 episodes, it should be noted, came from the three open-ended interviews of

Raymond, and centered around two drafts of a journal article that he was about to submit

for publication.

Table 11 of Appendix G outlines the breakdown of activities described by Raymond

about the writing of his journal article. As with the two proposal-writing projects

described earlier, the episodes centering on Raymond’s journal writing contained numerous

examples of rhetorical sensitivity (73 percent). On the context for re-writing the original

draft of the article, Raymond described the following:

Raymond: 1 The focus of the research is what happened when thesethings [proteins] were attached to membranes. 2 They had differentsurface chemistry. 3 But along the way, we saw something interesting,and because the focus was elsewhere, we did not, at that time wethought, you know, we saw something interesting and we did not look atit, we looked at it fairly extensively but, it, in itself, probably couldhave been something that maybe should have been studied as a whole. 4Rather than just, you know, there as chunk of an overall thing. 5 And so,I think, to give the reviewer some credit, I think the extent to whichwhat we looked at and the ideas we had, which were really good andinteresting—most people kind of, nobody really argued with theideas—but there was the gap between what we had and the ideas wehad. 6 It was probably kind of wide. 7 You know the conclusions and theclaims we were making were based on the data, then they were based on

124

my own editorial. 8 So that’s the original alignment. 9 And havinglooked at it for about six months, and having done a couple of otherexperiments last summer to maybe explore some of the positive comments,positive in the sense that did you think of this, did you think of that(Open-ended interview).

After explaining how their interest in protein composition evolved over time

(sentences 1 through 4), Raymond goes on to describe the reviewer’s response to his first

draft (sentence 5). As he had stated in other interviews, Raymond recognizes the

importance of couching his claims in his existing data; in sentence 5, for example, he

suggests that the reviewer’s problem with his article was that there was a “gap between

what we had and the ideas we had.” In sentence 7, Raymond further describes a major

shortcoming of his first draft when he says that although “the conclusions and the claims

we were making were based on the data,” they were also “based on my own editorial.”

Lastly, Raymond explains his strategy for responding to the reviewer’s comments; that is,

he identifies what he refers to as “positive comments,” comments which emphasize “did

you think of this, did you think of that.” These questions, in turn, appear to act as

heuristics for evaluating the first draft of his article.

Table 12 of Appendix G shows the breakdown of activities that Raymond described

as taking place during the planning stage of the article project (which accounted for seven

out of total 23 episodes). Five out of the seven, or 72 percent, of the planning episodes

contained some reference to rhetorical strategies used in writing the journal article; and

most of these references occurred in conjunction with references to the potential audience’s

reaction to the written article (58 percent):

Raymond: 1 So we went at it experimentally, we, you know, did somepartial scan experiments and all that. 2 They were convincing to us, butthen you need to, you have to show big bumps. 3 Your pictures have tohave big bumps. 4 So the bumps weren’t big enough. 5 So at that pointI’m there, well, what I can do is pare it down to a communication, send it

125

in somewhere and say, hey, there’s something interesting going on here,subsequent work will clarify this (Open-ended interview).

In this excerpt, Raymond describes how they carried out partial scan experiments,

but obtained disappointing results (sentences 1 through 3); interestingly, although the

results are “convincing” to him and his group, he is aware that the results will not convince

potential reviewers. His anticipation of the audience’s reaction to his pictures, in turn,

informs his (potential) strategy of using the results in a communication rather than in a

full-scale article (sentences 4 and 5).

Only one of the 23 episodes contained an explicit reference to Raymond’s revising

process:

Raymond: 1 And at first reading I figured that the differences between usand them [another group working in a similar field] were mainly due to,first, we treated the protein much differently and it’s kind ofinconsequential. 2 I should note that study when I go through and revisethis thing. 3 In fact, I went at this thing about three or four times, justkind of revise it, where eventually my thinking has changed totally. 4So now this is pretty much written. 5 All it needs is figure legends,figures, and if we get that final result we’re looking for, slap that in, andthen it’ll go out before the summer starts (Open-ended interview).

This excerpt, again, exemplifies Raymond’s awareness of how his research “fits into”

existing research in the field. Although he realizes that his treatment of the protein does

not differ significantly from the treatment of another group of researchers (sentence 1), he

also recognizes the importance of acknowledging similar work (sentence 2). Also, this

excerpt emphasizes how Raymond uses writing to inform his on-going research plans; in

sentence 3, he describes how his “thinking has changed totally” during his revision of the

journal article and, in sentence 5, he reveals that scientific writing and data collection are

taking place simultaneously.

126

The frequency of references to evaluating the journal article, made during the open-

ended interviews, are reported in Table 13 of Appendix G. Evaluative references accounted

for 14 out of the 23 episodes (61 percent). Again, references to alternative rhetorical

strategies for producing the document dominated Raymond’s discourse (78 percent), as well

as references to potential audience reactions to the article (64 percent). In particular,

Raymond explicated the important role that the review process played in his writing of

the second draft of the journal article:

Raymond: 1 Okay the enzyme we’re working with’s a pretty technical,useful enzyme. 2 It’s used for analytical applications, a lot of it’s sold. 3It’s an unstable enzyme, so anything we learn more about it might beuseful to people that attempt to come up with ways of stabilizing theirenzyme preparations. 4 And so that’s sort of our original mind-set; wentthrough, wrote it up, sent it in. 5 And, as it turned out, one review waspretty positive, the other was semi-hostile. 6 We probably could havejust revised it and sent it back in (Open-ended interview).

Beginning with a description of the enzyme he is studying (sentences 1 and 2),

Raymond goes on to describe why he feels other researchers will be interested in its study

(sentence 3). Although the reviews are both positive and negative (sentence 5), Raymond

and his collaborators decide to take a new approach to presenting their data. In the

following excerpt, he expands on the pragmatics and the usefulness of the reviewers’

feedback:

Raymond: 1 My sense was that what we did might come kind of close towhat somebody else was doing or they felt it was their area so it mighthave been an intrusion. 2 In fact, one of the comments said why don’t youuse NMR and when I read that that kind of tripped in my mind a verynarrow number of, a very small number of possible people that reviewermight have been.

Interviewer: NMR?

Raymond: 3 NMR being a technique that one can go in and assess what’sgoing in and assess what’s going on in the protein structure. 4 And thatreview was good, . . . it raised some interesting questions that we mighthave just been able to argue around and neglect, but it also gave an old

127

reference that we never found in the literature. 5 A very crude way ofdoing these kind of experiments too (Open-ended interview).

This excerpt is particularly illuminating in that it highlights the political nature

of the review process. In sentences 1 and 2, for example, Raymond suggests that one of the

reviewers might have taken a proprietorial stance against his article. As well, the

sentences reveal Raymond’s acute familiarity with a specialized group of researchers

doing similar work. Importantly, he is not particularly surprised by this proprietorial

element of the review process and, in sentence 4, goes on to describe the review as “good” in

that it provided him and his team with a reference that turned out to be very instrumental

in the subsequent re-write of the article. The review process, then, places researchers in a

context where they are asked to both contribute and conform to the existing beliefs, norms,

and expectations of their discourse community.

To conclude, I would like to examine two excerpts. In the first excerpt, Raymond

characterizes his re-write of the drafted journal article, revealing his awareness of both

his rhetorical strategies for presenting his findings and his potential audience’s reaction to

his argument:

Raymond: 1 So we’re attacking ourselves again, but by looking like we’reattacking ourselves, we’re leading up to the explanation I like the best. .. . 2 So by attacking ourselves first and using what other people havesaid to be aware of, we’ve tried to trash our own interpretation. 3 Whatwe do, we offered the bare minimum interpretation at the first, and bytrashing different alternatives kind of led to what I kind of like. 4 See,in this I was very careful when I said it. 5 If somebody were to read this,they’d have to go back and read again. . . . 6 A lot of things, keep itgeneral. 7 And as I kind of trash the alternatives, I get down tosomething specific, rather than beginning with a hypothesis and sayinghere’s something crazy that I’ve found. 8 I give the audience somethingto chew on, then I give them something, well, you know, we’ve looked itand it’s not an impurity and, in the beginning, everybody will probablysay, okay okay, and I could end at the domain thing and probably getpeople to buy it. 9 And not do anymore, just end it there. 10 But I trash it.

128

Interviewer: Why?

Raymond: 11 Because I don’t totally buy it either. 12 I’m trying to be kindof honest in this whole thing. 13 I don’t totally buy it either. 14 It worksbeautifully. 15 But, yeah, I don’t totally buy it. 16 So where we finallyend up now is we say basically that it’s probably, it could very well beboth things. 17 All right (Open-ended interview).

Raymond describes how he frames his argument or explanation in terms of

“attacking ourselves” in sentences 1 and 2. However, as he points out in sentence 3, his

original interpretations are predictably problematic. In sentences 4 and 5, Raymond refers

to the potential audience reaction to his line of argumentation and, in sentence 7, he

describes how he has deviated from the traditional problem-hypothesis-solution format

and, instead, started with a general overview of the issues and moved progressively

towards specific issues. Then, Raymond turns again to audience concerns (sentence 8), and

explains how he expects to extend his argument beyond the stage where they would

“probably . . . buy it.” His rationale for doing so, he posits, is that although his

explanation “works beautifully” (sentence 14), he does not “totally buy it either”

(sentences 11 and 13). He concludes the argument by suggesting that the domain and

heterogeneity perspectives towards enzyme structure are not necessarily mutually

exclusive notions (sentences 16 and 17). In another excerpt, Raymond explains that this is

often how he constructs a scientific argument: the “safe part” of the article is the

introductory, background material, the “core . . . that’s very difficult to attack,” and the

“speculative part” consists of the assertions that Raymond ultimately wants to make

regarding the structure of the enzyme.

In the next section, I elaborate on the 14 drafts that I collected of the collaborative

proposal. While the previous sections emphasized process information from the various

interviews and talk-aloud protocols, the next section emphasizes product information

129

integrated with process information from the taped meetings and numerous notes exchanged

between the two biochemical engineers over the duration of the project.

Managing the Collaborative Proposal Project

The Fourteen Proposal Drafts

As described earlier, the collaborative proposal that Raymond and Larry wrote

over a three-week period of time was actually the result of discussions they had dating

back to March, 1990. Several months later, Raymond wrote up a five-page prospectus, but

until the RFP was announced by NIH in September, 1990, Raymond and Larry had

basically continued to run separate experiments, consulting each other informally about the

relationship between their various results. Late that summer, they designed several

mutual experiments, which ultimately informed their initial collaborative effort.

Their proposal, in short, involved the characterization of a type of affinity

chromatography, IMAC, with two complementary goals: (1) as a means of purifying

proteins and (2) as an analytical tool for characterizing metal-binding proteins and

peptide-metal interactions. To accomplish this, they proposed to construct a family of

model proteins using site-directed mutagenesis and gene-fusion techniques. It was their

hope that, in removing bound metal ions from these proteins, the elution profile of the

proteins would provide them with information about zinc binding and non-binding proteins.

In addition, their goal was to assess whether proteins with multiple-binding sites utilized

particular sites or multiple sites. Finally, they intended to characterize the model

proteins and to contrast them with the metal-binding behavior of engineered versus native

130

proteins. To do so, they proposed to use numerous chemical engineering techniques: DSC,

light spectroscopy, CD spectroscopy, and metal-binding studies.

Table 14 of Appendix G presents the syntactical evolution of the 14 proposal drafts

(see, also, Appendix H for the chronology of the 14 proposal drafts). What is useful about

Table 14 is not only the differences across the drafts but also the similarities. Despite the

generation of over 500 sentences in three weeks (almost 10,000 words), Raymond and Larry’s

numerous drafts are extremely consistent in terms of word length—ranging from 1.6 to 1.8

syllables per word—sentence length—ranging from 19 to 25 words per sentence—passive

voice constructions—ranging from 25 to 37 per draft—and prepositions—ranging from 2.5 to

3.5 per sentence. The average paragraph length, with the exception the eighth and ninth

drafts, ranged between four and seven sentences per paragraph (drafts eight and nine

contained text that Raymond had downloaded as a computer file from Larry and that

contained no paragraph divisions; in draft 10, Raymond and Larry added paragraph

breaks and transitional sentences).

The development of the proposal, then, is marked by a consistency that suggests

that the biochemical engineers were employing a very well-learned set of writing

standards to the new text they created. That is, although the writing effort spanned

numerous days and drafts, and was a collaborative effort, the syntactical development of

the proposal was remarkably stable.

The finished proposal contained the following sections and subsections:

(1) Specific Aims;

(2) Significance (IMAC Background and Literature Survey, DSC Backgroundand Literature Survey, and Contributions of the Proposed Research);

131

(3) Preliminary Efforts (Zn (II)-IDA IMAC Experiments, Cu (II)-IDA IMACExperiments, Discussion of the IMAC Results, Overview of DSCExperiments, DSC Results, Discussion of DSC Results, and Summary andAnticipated Utility of DSC Studies);

(4) Proposed Research: Materials and Methods (Organism E. coli, ProteinIsolation, Protein Assays, Gel Electrophoresis, IMAC Chromatography,Site-Directed Mutagenesis, Gene-Fusion Techniques, and DSC);

(5) Research Plan (Choice and Construction of Model Proteins,Characterization of Model Proteins, IMAC Experiments, Metal-BindingStudies, and Collaborative Arrangements and Time-table)

(6) References; and an

(7) Appendix: Expanded Methods and Prior Work (Gene-Fusion Techniquesand Site-Directed Mutagenesis).

Table 15 of Appendix G shows the development, over the three-week writing

period, of the 14 drafts. The table is organized simply to represent the percentage, in

words, of each section of the proposal—the specific aims, significance, and so on—as it

evolved over the 14 drafts. Some sections were revised numerous times throughout the

process, however the changes were not substantial enough to increase the overall

percentage of that section. The Specific Aims section (see Appendix I), for example,

although it was altered continually over the three-weeks, generally made up less and less

of the total percentage of the proposal as more text was added to the remaining sections

(from 15 percent to 7 percent).

Table 15 reveals two particularly interesting aspects of the collaborative proposal

project. First, it points to a problem that Raymond and Larry encountered early in the

project. The first draft of the proposal consisted of numerous (previously drafted)

paragraphs cut and pasted into a single file. However, the majority of the text was

removed (excluding the Specific Aims section) from the second draft (almost 84 percent of

the original paragraphs). Also, 11 percent of the second draft’s paragraphs were new (a

132

first draft of the Significance section). In this respect, although the first draft gave the

two biochemical engineers a starting point for the project, very little of it was salvaged in

the second draft. The second trend revealed by Table 15 is the order in which Raymond and

Larry wrote the proposal. With the exception of the Research Plan (draft five) which

they started writing before the Proposed Research (draft seven), all of the other seven

sections of the proposal were completed in a linear fashion. Despite their linear revision

strategy, however, Raymond and Larry did revise the entire document thoroughly and

continually (unlike Selzer’s, 1983, study of the composing processes of corporate engineers

where the engineers revised little if at all).

Appendix J further elaborates on the percentage breakdown of the 14 draft

proposals, giving details of the Significance, Preliminary Efforts, and Research Plan

sections of the written proposal. In the next section, I take a detailed look at the evolution

of one section of the proposal, the Specific Aims section, and discuss how Raymond and

Larry revised and edited it over the 14 drafts.

Writing and Re-writing the Proposal’s Specific Aims Section

Appendix I shows the evolution of the Specific Aims section of the collaborative

proposal. The first draft, dated Monday, September 10th, consists of eight research aims.

The list also contains two notes that the authors intended to answer later, the first

attached to the fourth aim regarding crosslinking:

Is this protein crosslinked? If so we might want to look at reduced andoxidized forms to see if a variation in crosslinking pattern may haveoccurred due to the substitutions. A look at the structure could also maybeanswer this question (Note on first draft).

133

The second note follows the fifth aim and is a note from Raymond to Larry reminding him

“to look at finger literature to see what has been done on this topic or eliminate if you

think these experiments are not really feasible” (Note on first draft).

The second draft, dated Tuesday, September 11th, is changed substantially. The

first, second, and third research aims have been deleted, and two minor revisions have

been made (the addition of the missing noun, “cysteines,” to the seventh aim, and the

capitalization of the verb, “perform,” although they misspelled the word in the re-write).

In the third draft, completed two days later, the revisions are again quite local. In the

third research aim, they change the second sentence from “Determine, for example, if the

presence of the added fragment can suppress the effects the metal ions have on the

enzyme’s endotherm due to the fragment being able to bind the metal ion preferentially,”

to “Determine if the presence of the metal-binding, terminal fragment can suppress the

effects of the metal ions have on the enzyme’s endotherm due to the fragment being able to

preferentially bind the metal ion.” As Bazerman (1984, 1988) has pointed out, much of the

revising that scientists do is often limited to language refinement, which explains why the

biochemical engineers add the adjectives “metal-binding” and “terminal” to the noun

“fragment.” “Specificity,” in a scientific text is an important attribute. In aim four, they

extend the possible reasons for the existence of metal ion to include the secondary structure

formation and metal ion coordination and, in the fifth aim, they correct a minor

typographic error.

The Monday, September 17, draft of the Specific Aims section contained only two

local revisions—changing the typographical errors “difference amounts” to “different

amounts” and “Perfrom” to “Perform.” In the fifth draft, an interesting revision occurs: the

biochemical engineers re-incorporate the first three aims that they deleted from the

134

second draft. Due to confusions over which draft was the “final draft,” many of the text

changes made to first four drafts appear to have been lost or re-incorporated in the fifth

draft (for this reason, the typos corrected in the last version re-appear in this draft).

There were no changes made to the sixth, Wednesday, September 19, draft.

However, the seventh draft contains significant changes. The note following the fourth

aim has been deleted, as well as aim five (and its attached note). And aims seven and

eight have been deleted. In the eighth draft, Raymond and Larry add a new first aim.

The aim of constructing and producing a family of ß-lactamases, moreover, was a direct

result of a meeting held prior to the revision (see next section for more on the two meetings

held between Raymond and Larry). In aim five, they change “DSC” to “Differential

Scanning Calorimetry (DSC),” since this is the first time the term is used in the proposal

and, in aim six, they add the techniques they will be performing on the ß-lactamases,

“IMAC, CD spectroscopy, and binding studies,” to the DSC approach.

Drafts nine and 10 contain only superficial changes. Draft 11, however, is

significantly revised. A new second aim is inserted after the first aim. This aim, as well,

was discussed during the meeting between Raymond and Larry, and emphasizes the

possibility of “alternative ‘window’ opportunities” resulting from their research. In the

second aim (which has now become the third aim), the awkward phrase “which are

without any” is replaced by the phrase, “that are devoid of.” The fifth aim, now made

redundant by the addition of the new, second aim, is deleted, and the last aim re-written

significantly. The engineered ß-lactamases are now described as model proteins (to insure

consistency throughout the proposal), the vague noun, “fragment,” is replaced by “tail,”

and the word, “tail” is further elaborated as a “tail being able to bind the metal ion

preferentially” (a more technically appropriate definition of the phenomenon they are

135

describing). Draft 11 represents the most dramatic revision. Draft 12 contains an

additional, sixth aim, describing the IMAC procedure (which had been removed from the

fifth aim of the previous draft). And drafts 13 and 14 contain very few revisions: draft 13

contains a spelling correction, and draft 14 has the awkward phrase “exist via” replaced

by “can be generated by using.”

What the 14 drafts clarify about the revising habits of the two biochemical

engineers is that their strategies were varied and iterative. Some of the drafts were

revised significantly (e.g., drafts two, seven, 11, and 12) while others contained a few local

revisions (e.g., drafts three, four, six, nine, 10, 13, and 14). The only draft of the proposal

that was left completely untouched was the sixth draft. In addition, the only research

aims contained in the first draft of the proposal that still remained in the final draft were

aims one and two, although both were de-emphasized and moved to aims three and four.

As the next section will highlight, the various notes exchanged by the two biochemical

engineers and the two meetings they had to discuss the proposal were instrumental in

shaping the production of, not only the Specific Aims section, but the entire research

proposal as well.

Exchanging Notes to Aid Collaboration

Early in the proposal-writing process, the two biochemical engineers decided that

Larry’s IMAC experiments and results should dominate the proposal and that Raymond’s

DSC results should be used to support Larry’s findings. Hence the following note written by

Raymond and attached to the second draft of the Significance section:

Put in some DSC review stuff. However, seems that the main idea isattempting to resolve which part of the molecule is involved with themetal interaction and how the part can be altered? If this is the case,

136

then more significance should be placed on the IMAC component and theDSC review should mainly show competence? That is, I do not envisionproposing new theory although there may be room to do so in endothermdeconvolution. This would be of the form of proposing reversiblemicrostate models (Raymond note to Larry).

The note exemplifies a strategy that the two authors would follow in subsequent drafts.

Also attached to the same proposal were reminders written by Raymond to himself to carry

out various activities. These included reminding one of his graduate students to search the

relevant literature (“Elizabeth, put in our literature values [ref for lit values]”),

questioning the need for more data (“Would it be worth running water-water, buffer/salt-

buffer . . . to see if there is an exothermic CuSO4 ionization effect?”), and elaborating on

two possible hypotheses for explaining the same data:

Hypothesis 1:1 Me(II) binds to unfolded state thereby perturbing folded-

unfolded equilibrium toward the unfolded state. 2 The increasedavailability of His and Trp in the unfolded state would be the drivingforce.

Hypothesis 2:3 Me(II) binds to native state and has a mild chaotropic effect

thereby lowering Tm and DH.4 One can argue for #2: According to Smith the His is available

in chicken lysozyme. 5 Thus, unfolding lysozyme will offer no new His. 6Also, lysozyme has Trp, but RNAase does not. 7 Thus, if Trp binding wasimportant and resulted in Tm decrease, then the effect should be greaterfor lysozyme than RNAase because unfolding would offer a lot of Trponly for lysozyme. 8 (SH is not a factor for either). 9 Also, the greateramount of His in RNAase might allow for greater coordination numberand thus greater perturbation of native structure (10 Porvath shows thathigh His proteins bind more strongly to IMAC columns and postulatesthat His’s may be proximal thereby allowing for greater coordination). .. . 11 Does other evidence exist for ME(II) binding resulting in analteration of structure?. . . . 12 The main hole is does unfolding exposemore His in RNAase. 13 Bovine RNAase has a total of four and theBrookhaven databank should tell us where the Ne are.

The most notable aspect of the above note is also the most obvious one—that

writing had an on-going influence on Raymond and Larry’s process of interpreting their

137

experimental results. Also, it is interesting to note that Raymond leans towards

hypothesis two because he “can argue for” it (sentence 4). The note is actually a collapsed

and highly condensed (or abstracted) version of a complex argument. In it, Raymond

acknowledges connected research (sentences 4, 5, 10, and 11), identifies a potential, causal

relationship (sentence 7’s “if-then” structure), eliminates possible variables (sentence 8),

elaborates on his interpretation (sentence 9), and identifies a potential problem with his

interpretation (sentence 13).

Other notes—the following attached to the Specific Aims section of the sixth

draft—stressed which parts of the proposal required additional work:

1 Now have some clean up of results and link between two results sections.2 Appendix reduces length. 3 Next swat will expand proposed researchand clean up the text already there. 4 Will get feel for total length. 5Thus, trim later and return to front to tighten up if possible (Larry note toRaymond).

This note is particularly interesting in that, in it, Larry describes the multiple tasks that

have yet to be carried out; he refers to “cleaning” and “tightening” up the manuscript

(sentences 1, 3, and 5), addressing concerns about the length of the manuscript (sentences 2

and 4), building transitions (sentence 1), and elaborating on the existing text (sentence 3).

Figure 1 illustrates the significant changes that were made to the draft Larry is describing

(deleted text is stroked out and added text is underlined):

138

Figure 1: A snapshot of the additions and deletions made to the introductory paragraphs ofthe sixth draft. Underlined text represents text that was added to the existing draft andstroked out text represents text that was deleted from the existing draft.

Although the first paragraph of the proposal remains untouched, the second and

third paragraphs are re-written considerably. Transitional phrases (e.g., “however” and

139

“as a result”) are added to the text. Awkward phrases (e.g., “immobilized metal forms

additional coordination binding with appropriate solutes”) have been replaced by simpler

constructions (e.g., “the binding event”). The wordy “In an attempt to examine the

possibility of using. . .” is reduced to the phrase “To examine.” Their terminology has been

further refined to avoid misinterpretation; for instance, they replace the phrase, “few

proteins meet the structural requirement,” with “not all proteins have the amini acid

composition and structural requirements” and the word “metalloproteins” with

“metalloprotein content of B. coli.” They couch criticisms of existing research in a more

positive light. The sentence, “The IMAC as a tool for protein purification is only

understood in a very empirical sense,” for example, becomes, “However, despite the recent

promising technical developments in IMAC, its use as a tool for protein purification is,”

and so on. And finally, they have added references to issues discussed in other parts of the

proposal (the Methods and the Preliminary Efforts sections).

In the next section, I examine the two taped meetings held between the two

biochemical engineers and their influence on the plans made for writing the proposal and

on various drafts of the proposal itself.

The Two Taped Meetings

The biochemical engineers’ first meeting was held on September 21, 1990, for

approximately two hours. The conversation consisted of 212 turns, with the dialogue very

evenly distributed between Raymond (107 turns) and Larry (105 turns). Although I was not

present at the meeting, I collected copies of the notes made by both scientists. In general,

the focus of the meeting was on defining two particular enzymes in order to justify their use.

One of the enzymes was to have exposed His, whereas the other was to have two buried

140

ones. Larry agreed to bring out these differences in his discussion of the choice of model

proteins, and Raymond agreed to cover the topic in his DSC Discussion and the Binding-

Studies sections. The use of the two native proteins was a key aspect of the proposal and

helped define the uniqueness of their strategy. Without it, the work still looked good to

the two authors but, in Raymond’s words, had “the appearance of a fishing expedition

that lack[ed] a coherent attack plan” (IMAC Proposal Meeting Notes).

Having defined the two proteins, the authors would then be able to show how

physiochemical characterization and molecular biology would be more effectively

combined in their research. As well, during the meeting the authors generated numerous

new “angles” and experiments: (1) they established some uncertainty about their

analytical chemistry approach; (2) they added temperature as a variable for the IMAC

and binding experiments; (3) they discussed the possibility of using metal loading as a new

variable in the IMAC, binding, and DSC experiments; (4) they agreed that pH had to be

included in the use of DSC in the experimental plan; (5) they noted the use of AA as an

analytical tool, and; (6) they introduced an additional data-collection technique into the

proposal. Their goal for the next version of the proposal was to “clean up the master” and

to “put in additional/alternative section headings so integration [would be] possible and a

coherent picture [would] result” (Raymond and Larry’s notes).

The purpose of the second meeting, held on Wednesday, September 26 (five days

after the first meeting), was to take stock of the progress they had both made since they

last spoke. The second meeting had two things in common with the first: (1) the amount of

turns taken by Raymond and Larry were remarkably balanced (96 and 97 respectively), and

(2) the topics covered in the meeting emphasized the technical issues surrounding the

writing of the proposal and the work yet to be completed on the project. Table 16 of

141

Appendix G gives a breakdown of the planned tasks that Raymond and Larry established

during the two taped meetings.

In general, both meetings emphasized three broad types of task assignment: (1)

writing assignments centered around particular sections of the proposal, (2) writing

assignments dependent on the technical expertise of either Raymond or Larry, and (3)

discussions of administrative details. References to sections of the proposal made up 42

percent of the three assignments; references to writing based on technical expertise made up

40 percent of the assignments, and; references to administrative tasks made up the

remaining eight percent of the assignments. Importantly, all the assignments undertaken

by the two biochemical engineers were voluntary; that is, they were both clearly aware of

which sections of the proposal they each had the knowledge and skill to complete.

Interrater reliability, as I pointed out in Chapter 4, was not necessary for this part

of the data analysis since the assignment of tasks and the follow-up check to see whether

the tasks were, indeed, carried out were not difficult to interpret. An example will

reinforce my point. During the first meeting, while looking at page 21 of the proposal,

Raymond and Larry decided to add the possible effects of pH and tm to their Discussion

section:

Raymond: Ah, so I think then that this stuff would just add onto that.

Larry: PH effect, temperature effect. We can throw this in here.

Raymond: Yeah, down at the bottom. And start throwing in paragraphsthere. Then that section’s a little tighter. Right now, the way it stands,you know, this work could be a thousand miles apart. That’s okay, butit’s got to be a little more weaved together, in some sense (Meeting one).

In comparing the proposal draft dated before the meeting and the draft after the meeting, I

was therefore able to quickly establish whether or not the biochemical engineers had

142

altered that section of the proposal. It is notable that all the tasks established by

Raymond and Larry during the two meetings were incorporated into the written proposal,

further support for my argument that both meetings were held at crucial planning stages of

the proposal-writing project.

Finally, the breadth of tasks discussed in the two meetings deserves some

attention. During the first meeting, Raymond took responsibility for 15 tasks and Larry

took responsibility for 11 and, during the second meeting, Raymond took responsibility for

seven tasks and Larry took responsibility for 10; indeed, the distribution of tasks divided

between the two biochemical engineers is remarkably balanced: 22 tasks carried out by

Raymond and 21 tasks carried out by Larry. This distribution is generally consistent across

all three task types, supporting the notion that the collaboration was very much a joint,

evenly-shared endeavor.

Limitations of the Case Study

In presenting the results of my analysis of Raymond and his colleague’s writing in

this chapter, I made a deliberate rhetorical decision to move the tables to an appendix for

two reasons. Because data were collected using numerous methods and because I had

actively selected excerpts from the enormous amount of data available to me, I felt it

would be problematic to assume that the percentages I described were not (at least in part)

a reflection of my research interest and research questions.

Both these issues clearly raise concerns about the generalizability of my study of

proposal writing in biochemical engineering, and I believe I have addressed some of those

concerns in Chapters 1, 2, and 3. Moreover, the types of questions driving my exploratory

143

study of proposal writing (see Chapter 4), in general, necessitated my doing a detailed,

long-term case study of a single writer-scientist. I also recognize that it is impossible to

know which (if any) of Raymond and Larry’s writing strategies are peculiar to them as

individual writers, and do not necessarily reflect the writing practices of other

biochemical engineers. To this end, I feel strongly that my study compliments well

previous studies of the composing processes of academic and nonacademic writers in their

natural settings (e.g., Bazerman, 1988; Gilbert & Mulkay, 1984; Kaufer & Geisler, 1989;

Miller & Selzer, 1985; Myers, 1990; Rymer, 1988; Selzer, 1983, etc.).

In addition, because my study covered research activities and writing projects

spread out over two years, it is important to point out that my data do not represent a

“complete” picture of Raymond’s writing processes and products. Certainly, at least in

terms of the three writing projects, I was able to collect very detailed descriptions,

however, it would be naive to assume that my data do not omit numerous important reading

and writing events that shaped and influenced Raymond’s composing efforts (the NIH

review panel, e.g., has yet to inform Raymond and Larry of their proposal’s success or

failure at the time I am writing this chapter). As a result of the inevitable

“incompleteness” of my data, I have been very careful to avoid drawing causal conclusions

about the connection between the three writing projects (in fact, I collapsed and analyzed

them as a single statement about Raymond’s composing practices earlier in this chapter).

And I have also been careful not to characterize the collaboration between Raymond and

Larry as an ideal or representative one; indeed, they encountered numerous difficulties

during the project (e.g., keeping track of multiple versions of the proposal, hurredly

collecting data necessary for the proposal’s line of argumentation, wrestling with

conflicting hypotheses and how they should be presented in writing, and so on) that might

144

ultimately form the basis of what collaborating researchers need to be careful to avoid or

to pay special attention to.

And finally, because I deliberately took as much of an “insider’s stance” as was

possible for me (given my limited mathematical and scientific training), it was sometimes

difficult to separate myself from the biochemical engineers’ motivations and actions. For

instance, one colleague, while looking at an excerpt I had labelled “Audience

Characterization,” emphatically declared, “my gawd, lots of things are going on in here,”

and convinced me that audience construction was, indeed, interacting with the scientists’

perceptions of the field’s literature, their schema for the literature, and their implicit

sense of what a scientific contribution entailed. Because I had acquainted myself with the

research articles being discussed by the two biochemical engineers, my reading in part

interfered with my ability to interpret their discussion about the potential audience for

their research proposal.

These issues—the exploratory nature of the study, my active involvement in the

on-going data-collection process, and the small n—represent limitations that should not be

ignored and yet, in part, were inevitable given the nature of the broad questions driving my

study. In the next section, I outline the general conclusions of the study, and set the stage

for Chapter 6 and my discussion of the implications of the study for researchers interested

in scientific discourse.

General Conclusions and Remarks

The questions driving my study of a proposal writing in biochemical engineering

were, essentially, (1) how do scientists plan, write, revise, and evaluate proposals written

145

for research funding, (2) how do they characterize the intended audience for their

proposals, (3) in what way do they anticipate audience reactions to their texts, (4) what is

the interaction between proposal writing and scientific research, (5) how do they integrate

existing scientific research into their texts, (6) what role do discussions of technical issues

and constraints play in the proposal-writing process, and (7) how rhetorically sensitive

are proposal-writing scientists?

The most significant, and recurring, finding in the study was that writing—at least

the writing done by Raymond for the field of biochemical engineering—is rarely, if ever,

done in isolation. Collaboration plays a crucial role in the scientific and technical

inscription process (cf., Latour & Woolgar, 1979; Olsen, 1989).34 Even when Raymond

described writing projects that, for the most part, he had written in isolation, he referred

to a “we” that consisted of graduate students, colleagues, researchers that were part of his

discourse community, administrative assistants, and so on.

34 That we have tended to emphasize individual writing efforts has been discussed by

numerous researchers interested in the role of collaboration in writing. Fearingand Sparrow (1989), for example, argue that “instead of stressing the solitaryefforts of the writer to develop finely tuned prose, our textbooks and courses needto teach students about organizational and group processes—to plan and managecomplex organizational writing projects, negotiate with team members, resolvegroup conflict, and expect and deal effectively with the unexpected” (26). And, inpart, the lack of research on collaboration stems from our confusion about whatexactly it means to write collaboratively; for example, based on interviews with24 technical writers, Debs (1989) argues that most writers do not realize howmuch of their time is spent collaborating: “The majority of these writers at firstdenied—sometimes vehemently—that they engage in collaborative writing; later,after they had described the discussions and negotiations that go into the choicesthey make while writing different manuals, they concluded, often with a statementof surprise or reconsideration, that they do collaborate. In one case, of the eightytext features discussed during the interview (including questions of audience, style,and layout), sixty had been determined by the writer during discussions witheighteen other members of the organization, yet initially this writer too hadresponded that she rarely collaborates” (39).

146

And perhaps it is the growing amount of collaborative research and writing in

science that has resulted in my finding that rhetorical strategies play a large role in the

proposal-writing process. Collaboration, after all, heightens the amount of talk about

texts and, in some ways, makes it easier for researchers interested in scientific discourse to

“capture” strategies as they occur (cf., Doheny-Farina’s, 1986, discussion of corporate

writing and his emphasis on two extended meetings held by the writers involved).

In addition to revealing the integral role that collaboration plays in the writing

process of a biochemical engineer, the study also undermines the popular myth that

scientific and nonacademic writers spend little time and energy planning, revising, and

evaluating their texts (Broadhead & Freed, 1986; Selzer, 1983). The amount of planning

invested in all Raymond’s writing projects accounted for 44 percent of the episodes I coded,

and the amount of text evaluation infused into the writing process accounted for 38 percent

of the episodes.

Notably, my study corroborates Bazerman’s (1988) and Myers’ (1990) assertions

that scientific writing and scientific research are interdependent activities. As Raymond

puts it, they “map onto one another” with the proportions changing depending on the goals

of the research scientist. Also, it is important to note that Raymond (and his colleagues)

altered existing research plans and directions more in the proposal-writing process (27

percent of the episodes) than in the journal-writing project (approximately 5 percent of the

episodes). As Kennedy (1983) has emphasized, it is not surprising that “Nothing

substantial has ever got done without someone proposing what he would do, how he would

do it, what he would do with it, how much it would cost, and how long it would take to do”

(124). It is my hope, therefore, that the study supports and extends Myers’ (1990)

147

contention that proposal writing is one of the most significant types of writing that

scientists do.

The study also follows Bazerman’s (1988) strategy of balancing between a

description of scientific discourse as cognitively driven while at the same time being

contextually constrained. As Debra Journet argues in her (1990) review article,

[That] science as a social construct which is, at the same time accountableto empirical experience, is illustrated most vividly . . .in the long case study Bazerman offers of Arthur Holly Compton’sannouncement of what is now called the Compton effect. Here Bazermanshows not only the way Compton negotiates a knowledge claim, but alsohow Compton is constrained in that negotiation by such things as histheoretical commitments and existing research programs, the equipmenthe uses, the data he turns up (165-166).

Raymond and Larry, as well, faced numerous constraints during the proposal-

writing effort, the first being a shortage of time between receiving the RFP and the NIH

deadline for submission. Other constraints included their lack of familiarity with the

research that applied to their specific technical expertise, the RFP standards and

guidelines for submission, their awareness of the research interests of the (potential)

members of the review board, and so on; and all these constraints, in turn, motivated the

two biochemical engineers to find suitable solutions, to make and re-make research and

writing plans, and to revise and re-evaluate the proposal’s line of argumentation.

Finally, the study emphasizes the interaction between the writing processes of a

biochemical engineer and the texts that he creates (for publication and for research

funding). The business of science is, as Latour and Woolgar (1979) have argued, the business

of inscribing (or transcribing). Revising and re-thinking numerous drafts of scientific

writing, at least for Raymond, played a key role in his writing process as well as

influencing his research activities.

148

In the next, and final, chapter, I discuss the implications of my study of a

biochemical engineer and his proposal-writing process. First, I discuss the role that

alternative methodologies played in providing different “windows” into the nature of the

proposal-writing process. This leads to an examination of a re-occurring pattern in my data

(and in the data of other studies of scientific writers, e.g., Gilbert & Mulkay, 1984, Rymer,

1988); that is, I argue that scientists use a storytelling repertoire to describe and discuss the

science they are writing about. Moreover, I explore the implications that the metaphor of

storytelling has for researchers interested in scientific proposal writing. Finally, I posit

that scientists (at least the two biochemical engineers I studied) are far more rhetorically

sensitive than we, in rhetoric and sociology, have tended to acknowledge. This, in turn,

leads to a call for research on the relationship between a rhetoric of science and technology

as argument and a rhetoric of science and technology as narrative.

149

Chapter 6—Discussion and Implications

I think it’s going to be a classic study, and so I’m excited about writing it. .. . I have thought about it. I sort of held it, oh like a little jewel in theback of my head, uh, it . . . gives me a lot of pleasure just to think about it,and the time is right. . . . It’s a beautiful story (interview with Subject-Scientist S, 221).

Rymer, J. (1988). Scientific Composing Processes: How Eminent ScientistsWrite Journal Articles. Writing in Academic Disciplines: Advances inWriting Research, Vol. 2. D. A. Jolliffe (Ed.). Norwood, NJ: Ablex, 211-250.

Putting it into words makes you think about it more than just doing things(interview with Subject-Engineer, 280).

Winsor, D. A. (1989). An Engineer’s Writing and the CorporateConstruction of Knowledge. Written Communication, 6 (3), 270-285.

. . . I have heard a number of experimental psychologists say in responseto my chapter on the writing of their field, “the practices you describeare not rhetoric; they are simply good science” (320).

Bazerman, C. (1988). Shaping Written Knowledge: The Genre andActivity of the Experimental Article in Science. Madison, WI: The U ofWisconsin P.

At its most general level, this dissertation has been about discourse in science and

engineering. The first chapter focussed on the role that research proposals play in the

dissemination and construction of knowledge in science, and relied on an adaptation of

Bazerman’s (1988) model of the complex negotiation that takes place between scientists,

their writing, research, and the funding agencies that support them. It was argued that

researchers interested in the rhetoric of science and technology need to better understand

150

the relationship between scientists, scientific research, proposals for research funding, and

the agencies that fund them.

Chapter 2 expanded on this discussion, and drew on literature from cognitive

science, academic and nonacademic research, the sociology and history of science, and the

rhetoric of scientific and technical discourse. After briefly characterizing the

contemporary scientific proposal writer, I outlined why the study of science (theor ia ) and

the study of rhetoric (phronesis) have traditionally been separated. This lead to an

overview of the contemporary rhetorical interest in scientific discourse and, in particular,

highlighted the need for a rhetorical analysis of a genre that, to date, has received little

attention—the scientific research proposal.

In Chapter 3, I introduced two exploratory studies—the first, a protocol-based

study of fifteen technical and professional writing students as they composed short

proposals, and the second, an interview-based survey of the funding practices of fifteen

professional academics. Both studies represented useful starting points for my inquiry into

the general nature of the proposal-writing process, in that they showed how issues of

audience and tone influence the proposal-writing process, and how management and

organizational skills factor into the overall funding process. Importantly, both studies

also highlighted the need for a third study of proposal writing (described in Chapters 4

and 5), and suggested that the third study employ a myriad of data-collection techniques.

Using a case-study-based approach to studying writing in a scientific setting was,

for me, an important research decision. Rose (1985) has made a similar and convincing

argument for the same sort of departure from strictly qualitative or strictly quantitative

approaches to data collection. Describing the tension between writing researchers

151

interested in a case-study-based approach to data collection and those interested in more

quantitative approaches, he writes

The naturalistic camp champions interpretation, rich detail, “thickdescription,” while the more experimentally oriented camp insists onmeasurements, numerical analysis, the discussion grounded in statistics.The ideal text for the first group becomes the case study; for the second,it’s the research article with its attendant tables and charts. But whymust these be the two primary choices—two extremes pitted againsteach other? I would suggest that the most enlightening andcomprehensive writing about writing would fuse these two approaches,would weave statistics into descriptions and provide interpretive humancontexts for measurements. We in composing-process research need a wayto write about our findings that blends the interpretive and metaphoricwith the baldly referential and notational. How else will we render therichness of the writing act? (258).

As I described in Chapter 4, taking a Participatory Approach (PD) to data collection also

contributed to what I saw as an important need for multiple measures of the same writing

activities. The PD approach emphasizes the evolutionary and collaborative nature of

data collection, and stresses the need for studies of scientific writing that incorporate

scientific writers into the research process. In this way I have tried, in Chapter 5, to

“weave statistics into descriptions and provide interpretive human contexts for

measurement” in what I believe is a useful and convincing way (see Spilka’s, 1988, study of

six corporate engineers’ which bases its results on data collected using methodological

triangulation).

My goal for the third study, therefore, was to document, in detail, the proposal-

writing efforts of a professional biochemical engineer over an extended period of time. In

particular, the following questions and issues required further investigation: (1) how do

academic writers represent or characterize the intended audience for their proposals; (2)

how does the proposal-writing process influence academic research plans or goals; (3) how

do academics anticipate their audience’s response to their proposals; (4) how do the

152

discourse conventions of the field influence academic proposals for funding; and (5) how do

academic proposal writers integrate existing scientific research into their writing efforts?

Chapter 5, the results of the case study, confirmed many of my intuitions about the

role of proposals and proposal writing in scientific and technical settings. Indeed, my

findings undermined some of the myths that I believe we are currently guilty of

promulgating in the classroom, for example, “proposals are written before the research is

carried out.” Raymond, the biochemical engineer that my case study featured, and his

colleague-collaborator, Larry, for instance, were exceedingly audience-oriented. In

addition, their characterizations of the audience and anticipation of potential reactions

that audience might have to their writing influenced the amount of time and energy that

they spent weighing various rhetorical alternatives and problem-solving strategies. That

is, the biochemical engineers’ actively altered their research proposal’s subject-matter and

line of argumentation based on their sense of audience’s potential reaction to the text; thus,

they constantly referred to the need to “hedge,” “couch,” “omit,” “cover our butts,” “be

careful about,” “avoid getting nailed on,” and so on, throughout the planning,

writing/revising, and evaluation of their text.

My data (from open-ended interviews, discourse-based interviews, tape-recorded

meetings, and talk-aloud protocols) also support Greg Myers’ (1985b, 1990) assertion that

scientific research, journal-article writing, and proposal writing are highly

interdependent. In fact, the interaction between the writing process and the biochemical

engineers’ goals for the research was more intense in the proposal-writing project than it

was in the journal-writing project. A notable tension, between the feasibility (“safeness”)

of the existing data and the hypotheticality of their interpretations (“speculative

153

editorial”), surfaced in numerous episodes, and appeared to play a major role in the writing

of all three projects.

The results also indicated that the two biochemical engineers spent a considerable

amount of time incorporating existing research literature into their texts. Notably, at least

for Raymond, incorporating research literature into his writing happened most frequently

in the revising episodes. In the planning and evaluation stages of the projects, he tended to

refer abstractly to “the work being done by other researchers,” and to how his research

differed or was similar to that research. In addition, both Raymond and Larry were

careful to present the work of others in a positive rather than in a negative light; this

finding, too, is discussed in the extensive literature on reference- and citation-use in science

and engineering (see, e.g., Gilbert, 1977).

Finally, Chapter 5 reveals the complex and social nature of the proposal-writing

process. Not only were Raymond and Larry engaged in a large, collaborative writing

project, but they were also negotiating numerous tasks with graduate students, technicians,

enzyme sales representatives, existing texts and technological platforms, administrative

staff, and with the perceived panelists who would eventually review their proposal.35

These interactions took various forms—exchanged notes, conversations in the hallway and

laboratory, telephone calls, fax machine correspondence, formal and informal meetings,

35 In fact, the negotiation over the written proposal goes well beyond the confines of

the time-period that I studied. After being submitted to the initial reviewcommittee, the proposal would then be summarized by an NIH executive secretarywho would submit his recommendation for staff review. The staff reviewerswould then submit their recommendations to the national council of each institute,who would, in turn, submit a final report to the institute’s director (see Murphy &Dean, 1984, 1986). The proposal, of course, can be rejected at any level of theNIH review process, although the initial review committee is the crucial hurdle.

154

computer mail, borrowed articles, figures and tables—all surrounding and incorporated into

the proposal draft as it evolved across the 14 versions.

In this chapter, rather than re-iterating the results of Chapter 5, I want to forward

several perspectives towards scientific proposal writing and the study of scientific and

technical discourse that have emerged from my study of a biochemical engineer and his

composing processes. The first perspective is that different methodologies for studying

proposal writing in science highlight different aspects of the overall process. What all

methods have in common, however, is that they emphasize the constructive nature of

research in the rhetoric of science; that is, researchers and participants are constantly

attempting to build plausible stories to describe their objects of inquiry. In this regard, the

second perspective holds that it is fruitful to view scientific proposal writing as a type of

storytelling or narrative. The third perspective is of scientific writers as rhetorically

sensitive or self-conscious. I define rhetorical sensitivity as the awareness, on the part of

the two biochemical engineers, of the contingent nature of their data-collection techniques,

their data, claims, and interpretations. Indeed, I believe a view towards scientists as

storytellers and scientists as rhetorically sensitive might have significant ramifications

for the teaching of both scientific writing and scientific reading. Thus, students might

benefit from instruction that bases itself in situated science stories, rather than from

generalized, isolated facts and lists.

Methods as Windows on Proposal Writing

Throughout this dissertation, when the discussion called for it, I have cited the

numerous studies that have called into question certain methodologies and their strengths

and weaknesses in terms of providing writing researchers with useful data for documenting

155

the complex processes of composing.36 There can be no doubt that any methodological

approach can be characterized as foregrounding certain aspects of a given process while de-

emphasizing other aspects, and this claim deserves some attention, given my goal of

applying multiple data-collection techniques to the study of proposal writing in science

and engineering.

At the most minimal level, any methodology can be said to have strengths and

weaknesses, particularly depending on the research questions one is attempting to answer.

In addition, various researchers have claimed that certain methodologies are better suited

to specific types of writing and particular contexts within which that writing takes place.

Table 1 outlines one way of viewing the influence that different methodologies have on

both the nature of the data collected as well as on the kinds of implications that can be

drawn from those data:

36 Although these references are cited throughout the dissertation, a condensed list of

pertinent readings includes the following: Berelson, 1971; Brenner, Brown, andCanter, 1985; Brown and Herndl, 1986; Clifford, 1980, 1982; Cooper andHolzman, 1983; Doheny-Farina and Odell, 1985; Ericsson and Simon, 1980,1984; Garfinkel, 1967; Geertz, 1973, 1983; George, 1959; Gross, 1990a; Hayesand Flower, 1980; Mulkay, Potter, and Yearley, 1983; Newell and Simon, 1972;Odell, Goswami, and Herrington, 1983; Steinberg, 1986, and; Swarts, Flower, andHayes, 1984.

156

Methodology Emphasis onCognitive orSocial

Research(er)Intrusion

Ability toQuantify

Emphasis onProcess orProduct

Talk-aloudProtocols

cognitive high excellent process

Disc.-basedInterviews

cognitive-social high good process-product

Open-endedInterviews

social-cognitive high poor process

TapedSessions

social low poor process

WrittenDrafts

cognitive low excellent product

ExchangedNotes

cognitive-social low poor product-process

E-mailMessages

cognitive-social low poor product-process

Table 1: Rough comparison of alternative methodologies and the perspectives they havetraditionally supported.

The above table is, of course, a tentative comparison of alternative methods and

the potential strengths and shortcomings of each given our goal of producing useful

descriptions of writing in science and engineering. Thus, for example, I have characterized

talk-aloud protocols as focussing on cognitive rather than social issues despite the recent

emphasis on categorizing the effects that different task environments have on our writing

processes (cf., Flower, 1989). And, similarly, I have characterized taped sessions as

emphasizing social issues, despite our ability to analyze them for examples of individual

agency and intention.

My rough characterization of differences between methods, however, is important

in that it highlights how applying multiple data-collection techniques can aid our goal of

constructing plausible stories about scientific writing. Indeed, data collected using

157

different methods might uncover anomalies and conflicts in our representations of the

writing processes being studied (cf., Mulkay & Gilbert, 1984).

In the next section, I discuss the perspective towards proposal writing as

storytelling or narrative. My argument is that, just as rhetoricians of science translate

scientific behavior into stories for the members of their community, so too do scientists

employ a storytelling repertoire when describing scientific discourse and action.

Scientific Proposal Writing as Storytelling

Myers (1990) has stated that his major rationale for studying writing in biology

was to explicate the rhetorical nature of seemingly a-rhetorical scientific writing. In

doing so, he provides non-biologists with five strategies for reading and understanding

specialized texts: (1) Look for the rhetorical, (2) Reconstruct the social context, (3) Look for

related texts, (4) Look for the source of authority, and (5) Look for any links between

scientific language and everyday uses of language (255-258). In a sense, Myers is simply

asking non-biologists to “behave” like professional biologists when they read. Raymond,

for example, is reading the research articles of other scientists as arguments when he states

that he “doesn’t buy” a particular interpretation. And he regularly reconstructs the context

within which his research and the research of others takes place. Therefore, when he

disagrees with the same scientists’ conclusions, he is quick to add that it is because their

interpretation of the data have been “filtered by their research questions.” Similarly,

when Raymond receives the negative comments of one reviewer, he recognizes that the

comments might have had more to do with the reviewer’s personal research interests than

with his text. And finally, Raymond and Larry reveal an intense familiarity with other

texts and research activities of particular relevance to their research. That is, their

158

proposal is “intertextualized” by an existing body of literature and “contextualized” by

their knowledge of the field.

Similarly, Bazerman (1988) turns his study of scientific texts to pedagogical ends

when he argues, “As teachers, if we provide our students with only the formal trappings of

the genres they need to work in, we offer them nothing more than unreflecting slavery to

current practice and no means to ride the change that inevitably will occur in the forty to

fifty years they will practice their professions” (320).37 His recommendation, therefore,

is to support “rhetorical self-consciousness” in the sciences and engineering. To do so, he

recommends the following guidelines: (1) Consider your fundamental assumptions, goals,

and projects, (2) Consider the structure of the literature, the structure of the community, and

your place in both, (3) Consider your immediate rhetorical situation and rhetorical task,

(4) Consider your investigative and symbolic tools, (5) Consider the processes of knowledge

production, and (6) Accept the dialectics of emergent knowledge (323-329).

As I discussed at length in Chapter 5, the collaborative project appeared to

heighten the amount of time and energy that the two biochemical engineers spent

explicating their research assumptions and goals for the proposal. In addition, their

discussions of the expectations of a largely “chemical” audience shaped how they

anticipated their proposal’s place in the existing research literature and the biochemical

engineering community in general. Finally, both Raymond and Larry—perhaps as a result

37 Wells (1986), too, has argued that our goal, as teachers of scientific and technical

writing, should be “to work within the structures of technical discourse so thatstudents can negotiate their demands but also be aware of the limited but realpossibility of moving beyond them” (264). This argument, of course, points tothe need for an approach to teaching that stresses both general writing skills andparticular discourse skills. That is, instead of confining our subject-matter to thediscourse norms of a particular field (e.g., physics, mechanical engineering,biology, etc.), writing instructors need to emphasize how students can movebeyond certain constraints.

159

of their “engineering” perspective towards biochemistry—spent a sizable amount of time

describing, elaborating upon, and discussing the data-collection tools that they would be

using. Triangulating the results of the numerous methodologies, therefore, prompted

numerous discussions about potential difficulties or unanticipated problems that they

might encounter.

My argument is a variation on the themes of both Myers (1990) and Bazerman

(1988). That is, it is my contention that scientists and engineers already employ what

Gilbert and Mulkay (1984) refer to as a contingent or informal repertoire to describe their

scientific research and writing, and that one instantiation of this repertoire is their

tendency to characterize scientific writing as storytelling. When we emphasize the

storytelling repertoire and its use by scientists, moreover, I believe that we begin to

recognize that scientists are already “rhetorically self-conscious,” to use Bazerman’s (1988)

term.

Myers, in his (1990) chapter, “The Cnemidophorus File: Narrative, Interpretation,

and Irony in a Scientific Controversy,” writes the following about his use of the term,

narrative, to describe scientific discourse:

By narrative, I mean the selection and sequencing of events so that theyhave a subject, they form a coherent whole with a beginning and an end,

and they have a meaning that is conveyed by the sequence as a whole.38

If this seems an odd activity for scientists, it may be because we associatenarrative with storytelling, fictions, and falsehoods” (102).

Similarly, David Hamilton (1978) has pointed out that “Good writing . . . is always

carefully enacted and, in that sense, the qualification ‘scientific’ is hardly more

appropriate to writing physics than to writing stories” (32).

38 See also Freed and Roberts’ (1989) related discussion of “event sequencing” and

how we present narratives to describe our experiences.

160

When I argue that scientists use a storytelling repertoire to convey how their

research and writing evolve, I do not mean to imply that this strategy is peculiar to

scientists and engineers alone. Indeed, to one degree or another, we are all constantly

telling stories, creating and re-creating meaningful narratives. We only have to listen to

ourselves and to others as we describe a concert we have recently attended, a conversation

we have just taken part in, our rationale for applying to and attending a particular

graduate school, and so on, to realize that although each story consists of various “truths”

and “facts” (the concert was held at Carnegie Hall, the conversation lasted 15 minutes, and

the program we applied to was rhetoric), our opportunities for re-representing and re-

creating our stories are as plentiful as our imaginations allow.

What makes the storytelling metaphor so compelling when we are talking about

scientific discourse is that scientists have traditionally denied or hidden that aspect of

their process. This is not entirely surprising given its connection to rhetoric since, as Simons

(1989) has pointed out, “When ‘rhetoric’ is used in reference to scientists, textbook writers,

reporters, and the like, it is frequently a term of derision, a way of suggesting that they

have violated principles held high in their professions” (3). Hence the well-established

discourse conventions found in science and engineering (e.g., Gross, 1985), conventions which

actually reverse the sequence of events that many scientists go through when carrying out

scientific research. That is, scientific articles are generally designed to report, first,

existing research, an unanswered question (or questions), the methods used to answer the

question, the results of the study, and the implications for future researchers interested in

the question.

What my study and the studies of others (e.g., Bazerman, 1984; Collins, 1981a;

Latour & Woolgar, 1979; etc.) make clear, however, is that scientists rarely (if ever)

161

proceed in this fashion. Indeed, it would be difficult (if not impossible) to produce an

effective review of the literature without first having some research question or interest

which drove that review, and it would be problematic to develop meaningful hypotheses

without first having some data (even anecdotal or intuitive) to guide your investigation

(see Schriver, 1989c, on the various strategies that empirical researchers use to “invent”

new theories and practices in rhetoric and composition).

Importantly, I am not in any way, re-iterating the argument that scientific texts

are therefore deceptive or misleading. Scientific texts have a distinguished historical

tradition and their design very successfully “forces” scientist-writers to negotiate between

their desire to “speculate” and the scientific community’s desire to “maintain” the

validity of existing theory and practice.

Fisher (1984) discusses what he calls the “narrative paradigm” in a manner that

appeals to me for this very reason; that is, storytelling, in Fisher’s (1984) view, is less

about fraudulence than it is about finding “good reasons:”

The presuppositions that structure the narrative paradigm are: (1)humans are essentially storytellers; (2) the paradigmatic mode of humandecision-making and communication is “good reasons” which vary in formamong communication situations, genres, and media; (3) the productionand practice of good reasons is ruled by matters of history, biography,culture, and character along with the kinds of forces identified in the . . .language action paradigm; (4) rationality is determined by the nature ofpersons as narrative beings—their inherent awareness of narrativeprobability , what constitutes a coherent story, and their constant habitof testing narrative fidelity , whether the stories they experience ringtrue with the stories they know to be true in their lives . . . ; and (5) theworld is a set of stories which must be chosen among to live the good lifein a process of continual recreation (7-8).

162

The concept of “narrative probability,” or the search for coherence, is therefore not at odds

with our picture of scientific writing and research. And it does not turn the issue of how

scientist’s represent their research in writing into an ethical issue.

Journet (1990), too, points out how we tend to align storytelling with “fiction” and

scientific prose with “non-fiction:”

It may at first seem odd to associate narrative with science, as weusually think of narrative primarily in connection with storytelling andfiction. But theorists . . . now understand narrative more broadly as amode of interpretation common to factual as well as fictional writing. Inthe act of constructing narrative, a writer imposes temporal andsequential order on a mass of data by selecting and arranging“significant” events. The writer’s decisions about how significant eventsrelate to one another, are the product of a particular theoreticalorientation or conceptual framework. Narrative is thus a way ofconstructing knowledge that is important to any discipline that dependson historical explanation, including geology and evolutionary biology(164).

None of this changes the fact that, when we review the literature from the rhetoric of

science and technology, we still tend to be surprised when scientific writers are

characterized with traditional literary terms like “creative,” “imaginative,”

“resourceful,” “innovative,” and so on. Charles Bazerman, in his (1984) article, “The

Writing of Scientific Non-Fiction: Contexts, Choices, Constraints,” describes a scientist’s

writing process very much as though he were a fiction writer: “. . . by struggling with the

language the scientist writer can achieve a bit better fit between symbolization and

experienced world” (50).

Gilbert and Mulkay (1984), too, in their description of scientific writing,

emphasize the multiple choices that scientists have in representing “the reality” of their

research:

163

. . . the introductory sections of research papers, which often presentreviews of prior work and which to that extent are sources of historicaldata, can be seen to be rather finely crafted reconstructions in whichcertain kinds of events and actions are systematically excluded.

Thus, although we recognize that the history we constructed isbut one possible version of the history of the field, this “weakness”cannot be remedied by relying instead on other kinds of data. Otherversions of the history will remain both possible and plausible (34).

Similarly, as with scientific writing, numerous researchers have also alluded to

the constructive nature of scientific reading. Gragson and Selzer, for example, in their

(1990) article, “Fictionalizing the Readers of Scholarly Articles in Biology,” show how

skillfully two biological articles construct a presumed audience and point out that this

“reveals how self-consciously rhetorical are both performances (including the one that

seems completely ‘conventional’)” (30) [italics added]. Their argument is influenced by

Walter Ong’s (1975) stance that “The historian, the scholar or scientist, and the simple

letter writer all fictionalize their audiences, casting them in a made-up role and calling on

them to play the role assigned” (17).

Not only are scientific articles apparently “fictions,” but scientific graphics as

well. As Gilbert and Mulkay (1984) observe

It is not possible . . . to assess how far our respondents in general weretreating the fictional nature of pictures as intrinsic to the phenomena ofbioenergetics or intrinsic to the realm of pictorial discourse. What isclear, however, is that the great majority of them, that is, twenty out oftwenty-five, emphatically characterized pictorial representation intheir field up to the date of the interview as unavoidably speculative,hypothetical, uncertain, interpretative, highly personal, and so on (155-156).

164

As well, Myers (1990) has argued, not that scientific illustrations are conjectural, but that

they divert attention from the history which produced them and focus attention on “the

appearance and stories of the particular animals and plants studied” (159ff).39

So, too, is Raymond aware of the importance and usefulness of figures for

explaining his research (figures, i.e., with big bumps). During one open-ended interview,

for example, Raymond said that—given the time—he would have probably added more

figures; when I asked why, he said simply “to save words . . . people read and interpret the

pictures first.”

A closer examination of Jone Rymer’s (1988) study of Subject-Scientist J reveals how

extensively she employs the storytelling metaphor to describe his composing processes.

Seven out of the 27 excerpts that Rymer cites in the article (26 percent of the total cited)

refer to building a plausible story to describe the on-going research:

I usually take a most recent paper . . . so I’m making a start . . . the nextchapter of the story . . . you might as well say that we’re telling thisproject and previously we had this, and now we have such and suchavailable method (interview with Subject-Scientist J, 220) [italicsadded].

In the above excerpt, Subject-Scientist J characterizes each scientific paper that he

produces as another chapter in the book that is his research, with his “starting point” for

the next chapter rooted in his previous research project. In another excerpt, Subject-

Scientist J describes the pleasure he is having writing a current journal article:

This one is very special to me, and it’s not just because I want to addanother paper to my bibliography, but I think it’s very, very significant,and I’m pleased with the way it came out . . . . I think it’s going to be aclassic study, and so I’m excited about writing it . . . . I have thoughtabout it. I sort of held it, oh like a little jewel in the back of my head,

39 For extensive discussions of the role and representation of images in science, see

Knorr-Cetina and Amann (1990), Lynch (1985), and Lynch and Woolgar (1988).

165

uh, it . . . gives me a lot of pleasure just to think about it, and the time isright. . . . It’s a beautiful story (interview with Subject-Scientist J, 221)[italics added].

It might be argued that Subject-Scientist J is not aware of the potentially

controversial nature of the phrase “beautiful story” as a description for a scientific

research article, but I would extend another interpretation. I would argue that he is aware

of the creative element of the journal-writing process and, further, that he prides himself

on his ability to “tell a good story:”

The pieces all fit together, and they . . . were fragments at the beginning,and what’s interesting is that each section is almost an independent unitin itself. And it’s like in biology, uh, ontogeny. . . . And so each section isalmost like a microcosm of the entire piece. So within each section I . . .[am] looking for order, fitting it together, and then I’m really not surehow the various pieces are going to fit together, so I work on the pieces . .. and then I’m going to put it together (interview with Subject-Scientist J,226).

The next excerpt highlights the role that “sequencing” or organization plays in Subject-

Scientist J’s perception of scientific writing:

I guess I had outlined in my mind prior to drafting five or six fundamentalpoints I knew I was going to hit. And this was one of them. . . . I did notknow the relationship; I didn’t know the order or the sequence. I’m stillnot certain of the sequence. . . . It’s a starting point. There’s a logic to t h edevelopment of the story, and this just has to be in the middle or the end(interview with Subject-Scientist J, 231-232) [italics added].

And, finally, J’s description of his writing process highlights his motivation to create an

“interesting” or “engaging” story, to “sell” his ideas to his audience:

Just isn’t catchy. Gotta sell the stuff. Doesn’t mean that you gotta bedishonest. But it’s gotta be something that really catches people’s eyes,so they stand up and pay attention (interview with Subject-Scientist J,235).

166

This means, in effect, that he has to present a compelling narrative (or “drama”)

describing his process of discovery:

. . . all at once I got a further insight. . . . I was talking about the problemof heterogeneity . . . and then it suddenly occurred to me that the proteinswere not present in equivalent amounts due to variation in their content . .. I hadn’t thought about that before, and it just strengthened the dramaof the study and so I wrote it in. This is the first time since I’ve beenwriting this that something new appeared which I had not thought . . . Ihad thought that most of the stuff was already in my head (interviewwith Subject-Scientist J, 242) [italics added].

Rymer (1988) concludes her story/study with the following characterization of scientists

and scientific writing:

. . . scientist[s] are tellers of tales, creative writers who make meaningand who choose the ways they go about doing so. As Subject J muses,looking over his data and thinking about what it means to him: “I guessthe question I’m asking now is: How do you tell it? Do you tell it like itis, or do you tell it like you predicted, or the story it makes, or what?”(244).

As Rymer observes, her scientist “prepares himself for writing by immersing himself in his

story; what guides his writing, then, are his strong feelings and mental images about this

story he wants to tell” (232).

In conclusion, the storytelling metaphor for scientific and technical writing is

useful for several reasons. First, stories are always told by writers or rhetors and shared

with audiences or listeners. This emphasis, on the scientist-storyteller and his or here

scientific audience, in turn, highlights both the cognitive and the social dimensions of

scientific discourse production. Second, storytelling can be viewed as both a process- and a

product-oriented endeavor. The key is that all scientific texts must “place” themselves

within the stories that have gone on before them; they must adhere to Burke’s (1941, 1973)

notion of the “unending conversation” by refining, continuing, extending, altering, or

167

clarifying the narrative that currently exists. Third, they give writing instructors who

teach scientific and technical communication another means of describing scientific

research and writing. The narrative form is, after all, a well-learned one, and having

students view their data collection, interpretation, and report writing as storytelling

represents a potentially productive place to begin. Raymond shows this inclination often,

when he describes certain aspects of his writing as “partly fictional,” “speculative,”

“interesting,” and so on. In Fisher’s (1987) words, “There is no genre, not even technical

discourse, that is not constituted by both logos and mythos” (85).40

In the next section, I outline a perspective towards scientific and technical writers

as rhetorically sensitive, and argue that sociologists and rhetoricians have generally been

guilty of “objectifying” scientists and the work that they do.

Scientists and Engineers as Rhetors

Jeff Goldberg, in his (1988) “Anatomy of a Scientific Discovery: The Race to

Discover the Secret of Human Pain and Pleasure,” describes John Hughes, a Nobel Prize

recipient for his research into the nature of endorphins, as follows:

To Hughes the pig brains were an absolute necessity. He needed a lot ofthem, and here they were for free. He had tried explaining to thebutchers that he was looking for a chemical in those pig brains, achemical which resembled drugs derived from the opium poppy—anatural “morphine” produced from within the animal’s own body whichmight, someday, unlock the mysteries of safe relief for human pain. Afew of the butchers pretended to catch on, but Hughes quickly realizedthat gifts of whisky and a little money proved more effective tools ofdiplomacy than all his mad-sounding explanations (3) [italics added].

40 See, also, Steven B. Katz’s essay (1992) on narrative romance and the structure of

technical discourse. He recommends that, “In teaching students in science andtechnology how to communicate to various audiences, . . . perhaps we shouldteach them not only the language of experts and managers, but also the languageof public discourse, including the patterns of narration that permeate and constituteour culture” (400).

168

It is this kind of description of the scientist, striving for fame, bent on discovering

truths as yet unknown, and armed with “mad-sounding explanations,” that permeates

current, best-selling accounts of scientific activities. And it is also this type of

representation that has resulted in our contemporary fascination with the business or

“drama” of science (cf., Nelkin, 1987).

Unfortunately, this perception towards scientists, scientific reasoning, and

knowledge has not been without its shortcomings. That is, in glorifying scientists we have

inadvertently marginalized the knowledge that we produce in the humanities and social

sciences, so much so that I regularly encounter humanists who are flatly distrustful,

antagonistic, or insecure about the relationship between their liberal arts concerns and the

“vocational” concerns of “a-theoretical” empiricists, technologists, and scientists. Because

I have worked with and studied colleagues in the sciences and engineering, and because I

have taught technical communication for engineering and technology, I am regularly

surprised by this perception of scientists and what they do. Indeed, it seems to me that

academic scientists spend a considerable amount of energy trying to teach their students

about the contingency of knowledge and the relative strengths and weaknesses of the

methods they employ to collect data.

In short, I am more often struck by the similarities I have (as a writing researcher)

with scientists and engineers, than I am with the differences. And I believe this

development is going to continue, given recent, “creative” interests in science and

mathematics such as game theory, simulation modeling, new physics, and so on. Scientists,

that is, are more and more motivated to “imitate” nature rather than to claim that they

are in the business of “discovering” it, and imitation is traditionally the province of

artists, musicians, and creative writers.

169

Unfortunately, however, the die has been cast, and this, I believe, has resulted in

the portrayal of scientists and technologists as rhetorically naive or unsophisticated.

Researchers have claimed sociology, philosophy, and rhetoric as the ir “turf,” and

asserted that scientists and engineers have more to learn from us than we do from them.

Gross, for example, in his (1985) article, “The Form of the Experimental Paper: A

Realization of the Myth of Induction,” re-states a common conception towards

contemporary scientific knowledge-making when he writes: “The philosophy that acts as

scaffolding for the most sophisticated science is relatively naive, untouched by

contemporary developments in philosophy of science” (24). In his discussion of the

experimental research paper, a genre which “. . . satisfies a recurrent need to justify the

enterprise of experimental science in the face of the problematic nature of the inductive

processes on which that science relies for the creation and certainty of its knowledge” (16),

Gross (1985) further posits that “The philosophy of scientists is in fact at one with the

philosophy of the experimental paper, designed less to explore than to justify, designed to

perpetuate a myth, the myth of induction that will allow experimental science at all costs

to continue” (24).

Scientists, in this light, are not only philosophically naive, but they are also the

perpetuator’s of the myth of experimental ideology (a characterization made more

ominous by the phrase “at all costs”). As I described earlier in this section, I believe that

the development of this perception towards science is, in part, a reaction to our historical

deification of the scientific enterprise. But I also believe that it is an unproductive

characterization, and that a growing body of research in the sociology and rhetoric of

science and technology supports my stance. Many of the scientists interviewed in Gilbert

170

and Mulkay’s (1984) book, “Opening Pandora’s Box: A Sociological Analysis of Scientists’

Discourse,” for example, reveal an acute awareness of the contingency of their arguments.

Speaking of the discourse convention of using the third person in science, one

biochemist observes:

1 Everybody wants to put things in the third person. So they just say, “itwas found that.” 2 If it’s later shown that it was wrong, don’t accept anyresponsibility. 3 “It was found. I didn’t say I bel ieved it. It was found.”4 So you sort of get away from yourself that way and make it sound likethese things just fall down into your lab notebook and you report themlike a historian. . . . 5 Of course, everybody knows what’s going on. 6You’re saying, “I think.” 7 But when you go out on a limb, if you say “itwas shown that” or “it is concluded” instead of “we conclude,” it shouldbe more objective. 8 It sounds like you are taking yourself out of thedecision and that you’re trying to give a fair, objective view and that youare not getting personally involved. 9 Personally, I’d like to see the firstperson come back. 10 I slip into it once in a while. “We found.” 11 Eventhen I won’t say “I.” I’ll say “we” even if it’s a one-person paper. 12 Canspread the blame if it’s wrong [laughs]. [Leman, 57-8] (58-59).

The third person construction, according to this scientist, has evolved as a means of

obscuring personal responsibility for the findings and for maintaining an “objective” tone in

scientific texts. He also contrasts this perspective with terms that we, in rhetoric,

regularly employ to describe the “reality” of scientific research and writing—I believe, I

think, we conclude, we found, and so on. And these discussions of the difference between

actual scientific practice and scientific reporting permeate Gilbert and Mulkay’s (1984)

interviews with scientists.

Another scientist, criticizing the way science is taught in school, characterizes

actual scientific practice:

5 One is a myth, that we inflict on the public, that science is rational andlogical. 6 It’s appalling really, it’s taught all the way in school, thenotion that you make all these observations in a Darwinian sense. 7That’s just rubbish, this “detached observation.” “What do you see? 8Well, what do you see? God knows, you see everything. 9 And, in fact,

171

you see what you want to see, for the most part. 10 Or you see the choicesbetween one or two rather narrow alternatives. 11 That doesn’t getadmitted into the scientific literature. . . . [Spender, 32-3] (59-60).

Echoing sociologists and rhetoricians of science (Law & Williams, 1982; Mulkay & Gilbert,

1982a; Zappen, 1985; etc.), the scientist above is clearly aware of the differences between

scientific observation and the way it is presented in scientific literature; and, like

Raymond, he recognizes that interpretations and data are always, in part, a product of

what you are looking for. An ironic remark by another scientist reveals this perspective as

well:

Interviewer: 25 So the experimental evidence. . . .Barton: 26 At the end of the day solves everything [general laughter]. . . . [62-3] (96).

His laughter re-affirms, not that he is a “promulgator of the myth of the experimental

paper,” but rather the opposite; in effect, he is saying that experimental evidence does not

solve everything.

And finally, Cookson makes the most explicit statement of the contingent nature of

scientific knowledge:

. . . 11 Membranes are extremely complicated and it’s hard to know thatyou’ve ever got the variables all pinned down, so that when in fact youmake an observation that that observation is really what you think it is.. . . 30 What bothered me with this [alpha beta] episode was the finaland complete realization that there is no such thing as absolute truth. 31I mean, last summer, it really hit me like a ton of bricks that truth issimply what most people are willing to believe today. 32 And that’struth. 33 Tomorrow the population changes, people are not willing tobelieve the same stuff that they were willing to believe the day beforeyesterday, then truth changes. . . . [Cookson, 49] (104).

This excerpt, taken out of context, could well have been uttered by a sociologist or

rhetorician. In it, Cookson characterizes how undermining it was for him to realize that

172

“there is no such thing as absolute truth” and that “truth is simply what most people are

willing to believe today” (a worldview shared by most contemporary rhetoricians).

Certainly, not all sociologists and rhetoricians are guilty of portraying scientists

and engineers as rhetorically naive. Rymer (1988), for example, acknowledges the

rhetorical sensitivity of her scientists when she writes, “Subject K, echoing the

rhetoricians, social scientists, and philosophers of science, describes other scientists like

this:”

In writing papers, scientists go back and rewrite the history of theexperiment. . . . Eventually many of these scientists believe their ownrewriting and think that is the way they behaved, what they did.They don’t recognize the accidents, the intuitions, the leaps of faith, thesudden connections (interview with Subject-Scientist K, 241).

And Dorothy A. Winsor concludes her (1990) article, “Engineering

Writing/Writing Engineering,” with the following assertion:

Exertion of power through language is obviously not limited toengineers. As I worked on this paper, I was uncomfortably aware that I,too, was attempting to exert power. In particular, I am one of a group ofresearchers outside technology and science who claim that scientistshave no special way of knowing unavailable to the rest of us. It seems tome that in part we are reacting to the privileged position our cultureawards science and technology as ways of knowing. It is therefore likelythat we exaggerate the irrational aspects of science. As a scholar ofwriting, it is great fun to say that engineers are actually writing aboutother writing, a field I presumably know more about than they do. Theythink their field, their way of knowing is superior? Nonsense! Theirfield isn’t even their field; it is mine. But I also bow to privilegedscientific ideology by posing as knowing empirically with nothingbetween me and what I see. Unmediated knowledge, however, is notpossible for any of us. All writing, including mine, constructs the worldwhich the writer can bear to inhabit (169).

This passage is particularly interesting (and playful) for several reasons. First, Winsor

recognizes that knowledge in engineering, as in the humanities, is not outside the province

of non-engineers. Second, Winsor is aware of the historical reaction to the “privileged

173

position our culture awards science and technology.” Third, she acknowledges that it is

doubtful that engineers are naive enough to believe that “their way of knowing is

superior.” And finally, Winsor reminds us that she, too, is “posing” as a scholar who is

able to empirically identify “truths” and “facts” about writing in engineering. I believe

this is a critical move on her part, and one that makes engineers and technologists less

“objects” of inquiry, and more “participants” involved in the debate over how knowledge is

constructed in the academy. In Raymond’s words, “Rhetoricians, social scientists,

biochemical engineers, whoever; we’re all basically in the same business—trying to account

for uncertainty.”

Implications for Research and Teaching

More than anything, my study of the proposal-writing process of a contemporary

biochemical engineer has highlighted the importance of methodological issues in writing

research. In Chapter 2, I outlined some of the reasons that protocol-based and interview-

based studies of writing fall short of answering all the questions we have about writing in

the sciences and engineering, and made a case for a naturalistic study of proposal writing in

one particular discipline. In addition, I explored the implications of a Participatory

Design approach for researchers interested in studying the processes and products of

contemporary scientists.

What my study has also emphasized is the need for extended, long-term studies of

scientific and technical writers. Only by immersing ourselves in the on-going research,

reading, and writing of individuals outside our discipline can we hope to better understand

and appreciate the opportunities and constraints facing them as they go about constructing

knowledge for their own communities.

174

Also, along with making a case for the critical role that proposal writing plays in

the scientific construction of knowledge, I have also discussed the difficulties that

scientists face when attempting to manage large, collaborative proposal-writing projects.

Proposal writing, in effect, is research planning and, as such, is a difficult and time-

consuming activity for scientists. Since my data are from two biochemical engineers writing

a proposal for the NIH, it is my hope that other researchers will extend my analyses to

other disciplines and to other funding situations (e.g., corporate funding situations versus

federal ones); this includes collecting data on how scientists, technologists, and policy-

makers read and evaluate proposals for research funding.

As well, I believe that taking a perspective towards scientists and engineers as

storytelling rhetors, we emphasize the similarities between what we do in the humanities

and social sciences and what they do in the “hard” sciences and engineering; essentially,

we both apply methodological tools to “data”—whether those data are texts, talk, or

enzymes—and attempt to extrapolate theory that accounts for other texts, talk, or

enzymes. We have both experienced disciplinary controversies and tensions (whether in

the form of moving from Newtonian to Einsteinian physics or moving from structuralist to

post-structuralist theories of reading and writing) and we both recognize the contingency of

our knowledge claims.

Indeed, discussions of the constructivist and, therefore, contingent nature of

knowledge claims has gained the attention of researchers studying how to improve current

science teaching (e.g., Duit, 1991; Glynn, 1991; Glynn, Yeany, & Britton, 1991). Martin and

Miller (1988), in fact, recommend that scientific stories act as one source of information to

introduce students to the complexities of scientific thinking. Criticizing contemporary

science texts, they write

175

The science books adopted in most schools transmit a body of knowledgewith little attention to the bodies for whom that knowledge is intended.. . . These texts offer children no stories, no connections between forms andforces, between observers and observed. Without this profoundconnectivity which is the lifeblood of science, the body of scientificknowledge can be reduced to a corpse. Through a storytelling mode,scientific knowledge can be kept alive for children (259).

And Martin and Brouwer (1991), as well, urge that “an authentic portrayal of science would

be one that included, where appropriate, a development of the methodological,

epistemological, personal, private, public, historical, societal, and technological aspects

of science as well as addressing the question of the aims or purposes implicit within

science” (708).

My argument, therefore, is that extending our collection of case studies from

different disciplines has the potential to help us create a database of issues, concerns,

values, discourse cues, and so on. This database, in turn, should situate our teaching of

technical communication for the sciences and engineering, and result in a collection of

detailed descriptions of actual writing projects as they evolve over time. Technical

students are often skeptical about the importance of writing to their particular disciplines,

and this represents a useful way of letting them know that we, too, have collected and

analyzed our data. Our students need to learn that, as Raymond puts it, simply “presenting

the facts” is the “least interesting” aspect of most writing in the sciences and engineering.

176

Appendix A—Schriver’s Agenda of Research Questions

• What are the relationships among cognitive, social, and cultural factors in documentdesign?

• Are there differences between writers and readers across cultures? Do the same writingstrategies, for example, work for Japanese- and English-speaking audiences? Do readersin various cultures read and access information in similar ways?

• What is the relation between oral and written communication in social contexts? Forexample, how do oral and written language work together in educational, government, orcorporate contexts? Are there culture-specific patterns?

• What is the role of writers’ knowledge in document design? Subject-matter knowledge?Linguistic knowledge? Perceptual knowledge? Strategic knowledge? Rhetoricalknowledge?

• What constitutes skill in document design? What do document designers need to know tobecome expert? What experiences do they need?

• What is the best means for soliciting subject-matter knowledge from experts? Howshould writers without subject-matter expertise proceed?

• How can research on reading best be applied to document design?

• What are the principles underlying the visual design of effective text? Do some visualinformation structures meet readers’ needs better than others?

• What are the best strategies for designing texts that serve multiple functions, forexample, to inform and persuade?

• How does collaboration with other experts shape the nature of the document designprocess? What are the optimal collaboration points among people (e.g., writers,designers, and subject-matter experts) contributing to the same text?

• How can technology facilitate the document design process? What constraints doestechnology place on the document design process? What are the key features of a userinterface that would best support collaborative document design?

• How do the needs of expert audiences differ from those of lay audiences? How can theneeds of multiple audiences best be addressed?

177

Appendix A—Schriver’s Agenda, Continued

• Which text-evaluation methods are best suited for judging text quality? At whatpoint(s) in the document design process are particular text-evaluation methods mostuseful; e.g., what tests should be used for first drafts? Can we develop more sensitivetext-evaluation methods than are currently available? Are there effective combinationsof existing methods?

• What do writers learn from testing documents and observing readers interacting withtext? Are there long-term benefits? Can we consolidate this learning and teach it moredirectly?

• What are the most likely candidate text features for building online text critiquers? Canthe computer help us in text evaluation more than it has? (325).

178

Appendix B—Instructions for Taping a Protocol

A protocol is a technique whereby a writer orally records his/her thinking about a

document while writing it. You will be producing your “thinking aloud” protocol simply by

turning on your tape recorder and verbalizing your thoughts about the document. The most

important thing about this protocol is that you say everything out loud as you are thinking

and writing your message. (We realize, of course, that it’s impossible to say absolutely

everything you’re thinking while you’re writing, so just try to say as much as you can.) Be

assured that there are no incorrect or correct comments; stray remarks and seemingly

irrelevant comments are fine. The length of your commentary should be about thirty to

forty minutes. Because we would like your commentary to be as spontaneous as possible

under the circumstances, it’s especially important that you not “rehearse” your commentary

before recording it. If you are interrupted, simply note the interruption and its length when

you restart the tape (adapted from Blyler, 1989, 65-66).

179

Appendix C—Questions Asked of Fifteen Academic Researchers

• From whom do you seek research funding?

• How does dealing with industry differ from government funding agencies?

• What would motivate you to choose one funding agency over another?

• How do you usually communicate with funding agencies?

• When did you start working with funding agencies?

• What do you think funding agencies hope to gain from your research?

• How do you think your research will be used by funding agencies?

• How could funding agencies improve their relationship with you?

• What would you change about working with funding agencies?

• How do you think future academic-funding agency relationships will evolve?

180

Appendix D—The Episodes and the Multiple Data Types

Total # Total #Episodes

% of TotalWords

Max #Words/Episode

Min #Words/Episode

Mean #Words/Episode

Open-Ended Ints.

4 35 54.6% 482 58 270

Discs-Based Ints.

3 5 16% 214 55 134.5

Protocs. 2 11 73.2% 194 67 130.5

TapedSessions

2 20 24.3% 246 60 153

Tot/Ave 11 71 42% 284 60 172

Table: Breakdown and percentages of the 71 episodes analyzed across the multiple datatypes.

181

Appendix E—Questions Asked of the Biochemical Engineer

• What is the chronology of your research, journal writing and proposal writing?

• What is the relationship between your instrumentation and theory?

• Does the relationship between your instrumentation and your journal writing differfrom the relationship between your instrumentation and your proposal writing?

• When did you begin writing your proposal based on research you had carried out inthe laboratory and written about for a journal audience?

• What is the chronology of your proposal writing?

• What are your future research/funding plans based on this research?

• How has your research team evolved over the journal writing and proposal writingchronology?

• What were your reasons for collaborating on your proposal?

• Do you perceive any differences between planning your collaborative project andplanning the written proposal?

• How did you collaboratively agree on the organization of the proposal?

• What, if any, problems, difficulties, or tensions did you encounter in creating theproposal collaboratively?

182

Appendix F—Chronology of Raymond’s Writing Projects

Time Research activity or writing effort

1984-1987: Original laboratory work (focus on calorimetry techniqueand the unfolding of enzymes).

May, 1988: First draft of research article submitted to biotechnologyjournal.

July, 1988: Research article returned with reviewer comments (newfocus: concentration effect and purity issue).

September, 1988: More laboratory work and revision (armed withdata/answers; now attempting to resolve domain versusheterogeneity conflict).

End of May, 1989: Research article completed and re-submitted tobiotechnology journal.

September, 1989: Begins working on initial proposal-writing effort.

December, 1989: Research article published in biotechnology journal.

March, 1990: First discussions between Raymond and Larry about thepossibility of doing a collaborative proposal for NIH.

July, 1990: Raymond writes a short five-page outline of the proposal.

September 4, 1990: Raymond and Larry receive an RFP from the NIHinforming them that a special session of particularrelevance to their research is being held on October 1, 1990.

September 10, 1990: Raymond writes first draft of the proposal (based on hisJuly draft).

September 21, 1990: Raymond and Larry’s first meeting.

September 26, 1990: Raymond and Larry’s second meeting.

October 1, 1990: Proposal submitted to NIH.

183

Appendix G—Results of the Data Analysis

Planning Writing/

Revising

Evaluat-

ing

Total %

Open-ended

interviews

16% 1% 32% 49%

Discourse-based

interviews

1% 6% 7%

Protocols 16% 16%

Taped meetings 28% 28%

Total % 45% 17% 38% 100%

Table 1: Overview of the various data sources and their relation to planning, revising, andevaluation.

For the purposes of simplicity, in the tables that follow, I have abbreviated eachof the six categories as follows: “Char. Aud. (1)” are episodes where the biochemicalengineers characterized their audience; “Ant. Aud. (2)” are episodes where theyanticipated audience reactions; “Alt. Res. (3)” are episodes where they altered existingresearch plans; “Int. Sci. Res. (4)” are episodes where they integrated existing scientificresearch into their texts; “Ident. Tech. (5)” are episodes where they identified technicalissues and constraints; and, “Rhet. Alts. (6)” are episodes where they discussed rhetoricalalternatives or strategies for writing. In the tables that follow, these abbreviations willbe further shortened to “P” (for Planning), “R” (for Writing/Revising), and “E” (forevaluating) followed by the numbers one through six (e.g., “E3” indicates that, while“Evaluating” their texts, the biochemical engineers decided to alter their existingresearch plans).

184

Char.Aud. (1)

Ant. Aud.(2)

Alt. Res.(3)

Int. Sci.Res.(4)

Disc.Tech. (5)

Rhet.Alts.(6)

Total

Char.Aud. (1)

4% 4% 3% 1.5% 7% 19.5%

Ant. Aud.(2)

6% 1.5% 28% 35.5%

Alt. Res.(3)

7% 1.5% 8% 16.5%

Int. Sci.Res.(4)

10% 10%

Disc.Tech. (5)

3% 3%

Rhet.Alts.(6)

15.5% 15.5%

Total 4% 10% 7% 6% 1.5% 71.5% 100%

Table 2: Percentage breakdown of all three writing projects (i.e., the journal-writingproject, the individual proposal-writing effort, and the collaborative proposal-writingproject) and the various activities carried out by the biochemical engineer and hiscolleague.

185

P1 P2 P3 P4 P5 P6 Total

P1 3%

(6.5%)

3%

(6.5%)

1.5%

(3%)

1.5%

(3%)

4%

(9%)

12.5%

(28%)

P2 1.5%

(3%)

13%

(28%)

14%

(31%)

P3 1.5%

(3%)

1.5%

(3%)

6%

(13%)

9%

(19%)

P4 3%

(6.5%)

3%

(6.5%)

P5 3%

(6.5%)

3%

(6.5%)

P6 4%

(9%)

4%

(9%)

Total 3%

(6.5%)

4.5%

(9.5%)

1.5%

(3%)

3%

(6%)

1.5%

(3%)

33%

(72%)

46%

(100%)

Table 3: Percentage breakdown of the activities that the biochemical engineersemphasized during the planning phase of all three writing projects. P1 is “Characterizingthe audience,” P2 is “Anticipating audience reaction to the document,” P3 is “Alteringexisting research plans,” P4 is “Integrating existing scientific research or researchliterature,” P5 is “Discussing technical issues,” and P6 is “Discussing rhetoricalalternatives.”

186

R1 R2 R3 R4 R5 R6 Total

R1 1.5%

(9%)

1.5%

(9%)

R2 3%

(17.5%)

3%

(17.5%)

R3 1.5%

(9%)

1.5%

(9%)

R4 4%

(23.5%)

4%

(23.5%)

R5

R6 7%

(41%)

7%

(41%)

Total 1.5%

(9%)

15.5%

(91%)

17%

(100%)

Table 4: Percentage breakdown of the activities that the biochemical engineersemphasized during the writing/revising phase of all three writing projects. R1 is“Characterizing the audience,” R2 is “Anticipating audience reaction to the document,” R3is “Altering existing research plans,” R4 is “Integrating existing scientific research orresearch literature,” R5 is “Discussing technical issues,” and R6 is “Discussing rhetoricalalternatives.”

187

E1 E2 E3 E4 E5 E6 Total

E1 1%

(3%)

1%

(3%)

3%

(8%)

5%

(14%)

E2 4%

(10.5%)

1%

(3%)

13%

(35%)

18%

(48.5%)

E3 6%

(16%)

1%

(3%)

7%

(19%)

E4 3%

(8%)

3%

(8%)

E5

E6 4%

(10.5%)

4%

(10.5%)

Total 1%

(3%)

5%

(13.5%)

6%

(16%)

1%

(3%)

24%

(64.5%)

37%

(100%)

Table 5: Percentage breakdown of the activities that the biochemical engineersemphasized during the evaluation phase of all three writing projects. E1 is“Characterizing the audience,” E2 is “Anticipating audience reaction to the document,” E3is “Altering existing research plans,” E4 is “Integrating existing scientific research orresearch literature,” E5 is “Discussing technical issues,” and E6 is “Discussing rhetoricalalternatives.”

188

Char.Aud.(1)

Ant. Aud.(2)

Alt. Res.(3)

Int. Sci.Res.(4)

Disc.Tech. (5)

Rhet.Alts.(6)

Total

Char.Aud. (1)

6% 3% 3% 10% 22%

Ant. Aud.(2)

12% 15% 27%

Alt. Res.(3)

12% 3% 12% 27%

Int. Sci.Res.(4)

3% 3%

Disc.Tech. (5)

6% 6%

Rhet.Alts.(6)

15% 15%

Total 6% 15% 12% 3% 3% 61% 100%

Table 6: Percentage breakdown of the collaborative proposal-writing project and thevarious activities carried out by the biochemical engineer and his colleague.

189

P1 P2 P3 P4 P5 P6 Total

P1 4% 4% 4% 8.5% 20.5%

P2 8.5% 21% 29.5%

P3 4% 4% 17% 25%

P4 4% 4%

P5 8.5% 8.5%

P6 12.5% 12.5%

Total 4% 12.5% 4% 4% 4% 71.5% 100%

Table 7: Percentage breakdown of the activities that the biochemical engineersemphasized during the planning phase of the collaborative proposal-writing project. P1 is“Characterizing the audience,” P2 is “Anticipating audience reaction to the document,” P3is “Altering existing research plans,” P4 is “Integrating existing scientific research orresearch literature,” P5 is “Discussing technical issues,” and P6 is “Discussing rhetoricalalternatives.”

190

E1 E2 E3 E4 E5 E6 Total

E1 11% 11% 22%

E2 22% 22%

E3 33% 33%

E4

E5

E6 22% 22%

Total 11% 22% 33% 33% ~100%

Table 8: Percentage breakdown of the activities that the biochemical engineersemphasized during the evaluation phase of the collaborative proposal-writing project. E1is “Characterizing the audience,” E2 is “Anticipating audience reaction to the document,”E3 is “Altering existing research plans,” E4 is “Integrating existing scientific research orresearch literature,” E5 is “Discussing technical issues,” and E6 is “Discussing rhetoricalalternatives.”

191

Char.Aud.(1)

Ant. Aud.(2)

Alt. Res.(3)

Int. Sci.Res.(4)

Disc.Tech. (5)

Rhet.Alts.(6)

Total

Char.Aud. (1)

6.5% 6.5%

Ant. Aud.(2)

27% 27%

Alt. Res.(3)

6.5% 6.5% 13%

Int. Sci.Res.(4)

20% 20%

Disc.Tech. (5)

Rhet.Alts.(6)

33.5% 33.5%

Total 6.5% 93.5 100%

Table 9: Percentage breakdown of the initial proposal-writing project and the variousactivities carried out by the biochemical engineer.

192

R1 R2 R3 R4 R5 R6 Total

R1

R2 18% 18%

R3 9% 9%

R4 27% 27%

R5

R6 46% 46%

Total 100% 100%

Table 10: Percentage breakdown of the activities that the biochemical engineeremphasized during the writing/revising phase of the initial proposal-writing project. R1is “Characterizing the audience,” R2 is “Anticipating audience reaction to the document,”R3 is “Altering existing research plans,” R4 is “Integrating existing scientific research orresearch literature,” R5 is “Discussing technical issues,” and R6 is “Discussing rhetoricalalternatives.”

193

Char.Aud.(1)

Ant. Aud.(2)

Alt. Res.(3)

Int. Sci.Res.(4)

Disc.Tech. (5)

Rhet.Alts.(6)

Total

Char.Aud. (1)

4.5% 4.5% 9% 18%

Ant. Aud.(2)

4.5% 4.5% 51% 60%

Alt. Res.(3)

4.5% 4.5%

Int. Sci.Res.(4)

13% 13%

Disc.Tech. (5)

Rhet.Alts.(6)

4.5% 4.5%

Total 4.5% 9% 13.5% 73% 100%

Table 11: Percentage breakdown of Raymond’s journal-article project and the variousactivities he carried out.

194

P1 P2 P3 P4 P5 P6 Total

P1 14% 14% 28%

P2 58% 58%

P3

P4 14% 14%

P5

P6

Total 14% 14% 72% 100%

Table 12: Percentage breakdown of the activities that the biochemical engineeremphasized during the planning phase of the journal-article project. P1 is “Characterizingthe audience,” P2 is “Anticipating audience reaction to the document,” P3 is “Alteringexisting research plans,” P4 is “Integrating existing scientific research or researchliterature,” P5 is “Discussing technical issues,” and P6 is “Discussing rhetoricalalternatives.”

195

E1 E2 E3 E4 E5 E6 Total

E1 7% 7%

E2 7% 7% 50% 64%

E3 7% 7%

E4 14% 14%

E5

E6 7% 7%

Total 14% 7% 78% ~100%

Table 13: Percentage breakdown of the activities that the biochemical engineeremphasized during the evaluation phase of the journal-article project. E1 is“Characterizing the audience,” E2 is “Anticipating audience reaction to the document,” E3is “Altering existing research plans,” E4 is “Integrating existing scientific research orresearch literature,” E5 is “Discussing technical issues,” and E6 is “Discussing rhetoricalalternatives.”

196

1 2 3 4 5 6 7 8 9 10 11 12 13 14

WL 1.8 1.6 1.7 1.7 1.7 1.7 1.7 1.7 1.7 1.7 1.7 1.7 1.7 1.7

SL 20 20 25 23 22 23 22 25 25 19 19 19 19 19

PL 4 5 6 6 6 6 6 17 18 7 7 5 5 5

#S 234 117 104 114 300 299 364 459 481 755 759 748 744 775

PV 27 23 27 25 36 36 35 37 36 29 29 30 30 29

#PP 3.3 2.6 3.2 3.0 3.3 3.3 3.2 3.5 3.5 2.6 2.5 2.6 2.6 2.5

Table 14: The syntactical evolution of the 14 proposal drafts over the three-week writingperiod: WL is the average word length in syllables; SL is the average sentence length inwords; PL is the average paragraph length in sentences; #S is the number of sentences ineach draft; PV is the percentage of passive voice constructions in each draft, and; #PP isthe number of prepositions per sentence.

197

1 2 3 4 5 6 7 8 9 10 11 12 13 14

(1) 15 12 12 12 11 11 10 9 8 7 7 7 7 7

(2) 25 4 4 4 19 19 15 22 22 20 21 21 22 20

(3) 8 84 76 74 50 50 40 32 33 24 24 22 21 22

(4) 8 8 7 6 6 6 7 6

(5) 32 20 20 27 29 30 27 27 28 28 30

(6) 5 5 5 6 6

(7) 10 10 11 9 9

Table 15: The development and percentage breakdown of each section of the 14 proposaldrafts over the three-week writing period: (1) is the Specific Aims section; (2) is theSignificance section; (3) is the Preliminary Efforts section, (4) is the Proposed Research:Materials and Methods section; (5) is the Research Plan section; (6) is References section,and; (7) is the Appendix: Expanded Methods and Prior Work section.

198

Meeting One Meeting Two

Raymond’sTask Plan

Larry’sTask Plan

Raymond’sTask Plan

Larry’sTask Plan

Section ofProposal

(6) (2) (4) (6) (18)

Specific Aims

1 2 3

Signific. 2 2 Methods 1 1 2 Research Plan

1 2 3

Tables 1 1 2 Figures 1 2 3 Refs. 1 1 1 3TechnicalExpertise

(4) (5) (4) (4) (17)

CD 1 1 DSC 2 2 4 Gene Fusion

1 1

Tail Selection

1 1

Metal Binding

1 1 2

pH/tm. 1 1 2 Site-dir. Mut.

1 1

IMAC 1 1 2 Model Choice

2 1 3

Admin.Details

(5) (3) (8)

Fact- finding

3 3 6

Farming work out

2 2

Total Tasks 15 7 11 10 43

Table 16: The distribution of planned tasks established during the two taped meetingsbetween the biochemical engineers.

199

Appendix H—Chronology of the 14 Proposal Drafts

Number Day Date # of Words

Version 1: Monday September 10th 5313 Words

Version 2: Tuesday September 11th 2440 Words

Version 3: Thursday September 13th 2624 Words

Version 4: Monday September 17th 2721 Words

Version 5: Tuesday September 18th 7018 Words

Version 6: Wednesday September 19th 7164 Words

Version 7: Thursday September 20th 8585 Words

Meeting 1: Friday September 21st

Version 8: Wednesday September 26th 11771 Words

Meeting 2: Wednesday September 26th

Version 9: Thursday September 27th 12308 Words

Version 10: Friday September 28th 14832 Words

Version 11: Saturday September 29th 14898 Words

Version 12: Saturday September 29th 14814 Words

Version 13: Sunday September 30th 14837 Words

Version 14: Monday October 1st 15272 Words

200

Appendix I—Evolution of the NIH Proposal’s Specific Aims Section

The First Draft—Monday, September 10th Version

1. To demonstrate that the portion of elution profile which are without anycontaminant proteins could be utilized for one step purification of target proteinfrom a cell extract by directing the elution of a target protein towards those areaemploying gene fusion techniques.

2. To assess whether a protein demonstrating multiple binding site for a metal ionwill utilize one site preferentially or multiple attachment will occur; and examinethe implications of these results for protein purification strategies whenemploying metal binding affinity peptides.

3. To examine a possible role for IMAC as a simple and rapid tool for identification ofzinc and other metal binding proteins in different preparations.

4. Contrast engineered ß-lactamases (i.e. contain same terminal sequence butdifference amounts of background histidine have been removed) to wild type withDSC to ensure the engineered protein is not conformationally heterogeneous and tocharacterize any alteration of conformation, or variation in crosslinking pattern,that may of occurred. (Is this protein crosslinked? If so we might want to look atreduced and oxidized forms to see if a variation in crosslinking pattern may haveoccurred due to the substitutions. A look at the structure could also maybe answerthis question.)

5. Characterize the transition of the terminal peptide in the presence and absence ofmetal ion to confirm whether finger structure formation can occur in the presence ofmetal ion and what the enthalpy change is compared to the protein (need to lookat finger literature to see what has been done on this topic or eliminate if you thinkthese experiments are not really feasible).

6. Contrast metal binding behavior of engineered proteins and native protein withDSC. Determine, for example, if the presence of the added fragment can suppressthe effects the metal ions have on the enzyme’s endotherm due to the fragmentbeing able to bind the metal ion preferentially.

7. Perform sulfhydryl group titrations with DTNB and 4-PDS to ensure that all theadded are present in reduced form. Additionally, determine if in the presence ofmetal ion, the kinetics of the probe reaction is altered which could be an indicationof reduced accessibility of the probe for cysteine due to secondary structureformation.

8. perform uv and near visible light spectroscopy experiments employing the bindingof Co(II) as a probe a Me(II)-S coordination occurring.

201

Changes to the First Draft—Tuesday, September 11th Version

1 To demonstrate that the portion of elution profile which are without anycontaminant proteins could be utilized for one step purification of target proteinfrom a cell extract by directing the elution of a target protein towards those areaemploying gene fusion techniques.

2. To assess whether a protein demonstrating multiple binding site for a metal ionwill utilize one site preferentially or multiple attachment will occur; and examinethe implications of these results for protein purification strategies whenemploying metal binding affinity peptides.

3. To examine a possible role for IMAC as a simple and rapid tool for identification ofzinc and other metal binding proteins in different preparations.

4. 1 . Contrast engineered ß-lactamases (i.e. contain same terminal sequence butdifference amounts of background histidine have been removed) to wild type withDSC to ensure the engineered protein is not conformationally heterogeneous and tocharacterize any alteration of conformation, or variation in crosslinking pattern,that may of occurred. (Is this protein crosslinked? If so we might want to look atreduced and oxidized forms to see if a variation in crosslinking pattern may haveoccurred due to the substitutions. A look at the structure could also maybe answerthis question.)

5. 2 . Characterize the transition of the terminal peptide in the presence and absence ofmetal ion to confirm whether finger structure formation can occur in the presence ofmetal ion and what the enthalpy change is compared to the protein (need to lookat finger literature to see what has been done on this topic or eliminate if you thinkthese experiments are not really feasible).

6. 3 . Contrast metal binding behavior of engineered proteins and native protein withDSC. Determine, for example, if the presence of the added fragment can suppressthe effects the metal ions have on the enzyme’s endotherm due to the fragmentbeing able to bind the metal ion preferentially.

7. 4 . Perform sulfhydryl group titrations with DTNB and 4-PDS to ensure that all theadded cysteines are present in reduced form. Additionally, determine if in thepresence of metal ion, the kinetics of the probe reaction is altered which could bean indication of reduced accessibility of the probe for cysteine due to secondarystructure formation.

8. 5 . perform Perfrom uv and near visible light spectroscopy experiments employing thebinding of Co(II) as a probe a Me(II)-S coordination occurring.

202

Changes to the Second Draft—Thursday, September 13th Version

1. Contrast engineered ß-lactamases (i.e. contain same terminal sequence butdifference amounts of background histidine have been removed) to wild type withDSC to ensure the engineered protein is not conformationally heterogeneous and tocharacterize any alteration of conformation, or variation in crosslinking pattern,that may of occurred. (Is this protein crosslinked? If so we might want to look atreduced and oxidized forms to see if a variation in crosslinking pattern may haveoccurred due to the substitutions. A look at the structure could also maybe answerthis question.)

2. Characterize the transition of the terminal peptide in the presence and absence ofmetal ion to confirm whether finger structure formation can occur in the presence ofmetal ion and what the enthalpy change is compared to the protein (need to lookat finger literature to see what has been done on this topic or eliminate if you thinkthese experiments are not really feasible).

3. Contrast metal binding behavior of engineered proteins and native protein withDSC. Determine, for example, if the presence of the added metal-binding,terminal fragment can suppress the effects the metal ions have on the enzyme’sendotherm due to the fragment being able to preferentially bind the metal ionpreferentially.

4. Perform sulfhydryl group titrations with DTNB and 4-PDS to ensure that all theadded cysteines are present in reduced form. Additionally, determine if in thepresence of metal ion, the kinetics of the probe reaction is altered which could bean indication of reduced accessibility of the probe for cysteine due to secondarystructure formation and metal ion coordination .

5. Perfrom uv and near visible light spectroscopy experiments employing the bindingof Co(II) as a probe a of Me(II)-S coordination occurring.

203

Changes to the Third Draft—Monday, September 17th Version

1. Contrast engineered ß-lactamases (i.e. contain same terminal sequence butdifference different amounts of background histidine have been removed) to wildtype with DSC to ensure the engineered protein is not conformationallyheterogeneous and to characterize any alteration of conformation, or variation incrosslinking pattern, that may of occurred. (Is this protein crosslinked? If so wemight want to look at reduced and oxidized forms to see if a variation incrosslinking pattern may have occurred due to the substitutions. A look at thestructure could also maybe answer this question.)

2. Characterize the transition of the terminal peptide in the presence and absence ofmetal ion to confirm whether finger structure formation can occur in the presence ofmetal ion and what the enthalpy change is compared to the protein (need to lookat finger literature to see what has been done on this topic or eliminate if you thinkthese experiments are not really feasible).

3. Contrast metal binding behavior of engineered proteins and native protein withDSC. Determine if the presence of the metal-binding, terminal fragment cansuppress the effects the metal ions have on the enzyme’s endotherm due to thefragment being able to preferentially bind the metal ion.

4. Perform sulfhydryl group titrations with DTNB and 4-PDS to ensure that all theadded cysteines are present in reduced form. Additionally, determine if in thepresence of metal ion, the kinetics of the probe reaction is altered which could bean indication of reduced accessibility of the probe for cysteine due to secondarystructure formation and metal ion coordination.

5. Perfrom Perform uv and near visible light spectroscopy experiments employing thebinding of Co(II) as a probe of Me(II)-S coordination occurring.

204

Changes to the Fourth Draft—Tuesday, September 18th Version

1 . To demonstrate that the portion of elution profile which are without anycontaminant proteins could be utilized for one step purification of target proteinfrom a cell extract by directing the elution of a target protein towards those areaemploying gene fusion techniques.

2 . To assess whether a protein demonstrating multiple binding site for a metal ionwill utilize one site preferentially or multiple attachment will occur; and examinethe implications of these results for protein purification strategies whenemploying metal binding affinity peptides.

3 . To examine a possible role for IMAC as a simple and rapid tool for identification ofzinc and other metal binding proteins in different preparations.

1. 4 . Contrast engineered ß-lactamases (i.e. contain same terminal sequence but differentdifference amounts of background histidine have been removed) to wild type withDSC to ensure the engineered protein is not conformationally heterogeneous and tocharacterize any alteration of conformation, or variation in crosslinking pattern,that may of occurred. (Is this protein crosslinked? If so we might want to look atreduced and oxidized forms to see if a variation in crosslinking pattern may haveoccurred due to the substitutions. A look at the structure could also maybe answerthis question.)

2. 5. Characterize the transition of the terminal peptide in the presence and absence ofmetal ion to confirm whether finger structure formation can occur in the presence ofmetal ion and what the enthalpy change is compared to the protein (need to lookat finger literature to see what has been done on this topic or eliminate if you thinkthese experiments are not really feasible).

3. 6. Contrast metal binding behavior of engineered proteins and native protein withDSC. Determine , for example, if the presence of the added metal-binding,terminal fragment can suppress the effects the metal ions have on the enzyme’sendotherm due to the fragment being able to preferentially bind the metal ionpreferentially .

4. 7. Perform sulfhydryl group titrations with DTNB and 4-PDS to ensure that all theadded cysteines are present in reduced form. Additionally, determine if in thepresence of metal ion, the kinetics of the probe reaction is altered which could bean indication of reduced accessibility of the probe for cysteine due to secondarystructure formation and metal ion coordination.

5. 8 . Perform perform uv and near visible light spectroscopy experiments employing thebinding of Co(II) as a probe of a Me(II)-S coordination occurring.

205

Changes to the Fifth Draft—Wednesday, September 19th Version

1. To demonstrate that the portion of elution profile which are without anycontaminant proteins could be utilized for one step purification of target proteinfrom a cell extract by directing the elution of a target protein towards those areaemploying gene fusion techniques.

2. To assess whether a protein demonstrating multiple binding site for a metal ionwill utilize one site preferentially or multiple attachment will occur; and examinethe implications of these results for protein purification strategies whenemploying metal binding affinity peptides.

3. To examine a possible role for IMAC as a simple and rapid tool for identification ofzinc and other metal binding proteins in different preparations.

4. Contrast engineered ß-lactamases (i.e. contain same terminal sequence butdifference amounts of background histidine have been removed) to wild type withDSC to ensure the engineered protein is not conformationally heterogeneous and tocharacterize any alteration of conformation, or variation in crosslinking pattern,that may of occurred. (Is this protein crosslinked? If so we might want to look atreduced and oxidized forms to see if a variation in crosslinking pattern may haveoccurred due to the substitutions. A look at the structure could also maybe answerthis question.)

5. Characterize the transition of the terminal peptide in the presence and absence ofmetal ion to confirm whether finger structure formation can occur in the presence ofmetal ion and what the enthalpy change is compared to the protein (need to lookat finger literature to see what has been done on this topic or eliminate if you thinkthese experiments are not really feasible).

6. Contrast metal binding behavior of engineered proteins and native protein withDSC. Determine, for example, if the presence of the added fragment can suppressthe effects the metal ions have on the enzyme’s endotherm due to the fragmentbeing able to bind the metal ion preferentially.

7. Perform sulfhydryl group titrations with DTNB and 4-PDS to ensure that all theadded are present in reduced form. Additionally, determine if in the presence ofmetal ion, the kinetics of the probe reaction is altered which could be an indicationof reduced accessibility of the probe for cysteine due to secondary structureformation.

8. perform uv and near visible light spectroscopy experiments employing the bindingof Co(II) as a probe a Me(II)-S coordination occurring.

206

Changes to the Sixth Draft—Thursday, September 20th Version

1. To demonstrate that the portion of elution profile which are without anycontaminant proteins could be utilized for one step purification of target proteinfrom a cell extract by directing the elution of a target protein towards those areaemploying gene fusion techniques.

2. To assess whether a protein demonstrating multiple binding site for a metal ionwill utilize one site preferentially or multiple attachment will occur; and examinethe implications of these results for protein purification strategies whenemploying metal binding affinity peptides.

3. To examine a possible role for IMAC as a simple and rapid tool for identification ofzinc and other metal binding proteins in different preparations.

4. Contrast engineered ß-lactamases (i.e. contain same terminal sequence butdifference amounts of background histidine have been removed) to wild type withDSC to ensure the engineered protein is not conformationally heterogeneous and tocharacterize any alteration of conformation, or variation in crosslinking pattern,that may of occurred. (Is this protein crosslinked? If so we might want to look atreduced and oxidized forms to see if a variation in crosslinking pattern may haveoccurred due to the substitutions. A look at the structure could also maybe answerthis question.)

5. Characterize the transition of the terminal peptide in the presence and absence ofmetal ion to confirm whether finger structure formation can occur in the presence ofmetal ion and what the enthalpy change is compared to the protein(need to look atfinger literature to see what has been done on this topic or eliminate if you thinkthese experiments are not really feasible).

6. Contrast metal binding behavior of engineered proteins and native protein withDSC. Determine, for example, if the presence of the added fragment can suppressthe effects the metal ions have on the enzyme’s endotherm due to the fragmentbeing able to bind the metal ion preferentially.

7. Perform sulfhydryl group titrations with DTNB and 4-PDS to ensure that all theadded are present in reduced form. Additionally, determine if in the presence ofmetal ion, the kinetics of the probe reaction is altered which could be an indicationof reduced accessibility of the probe for cysteine due to secondary structureformation.

8. perform uv and near visible light spectroscopy experiments employing the bindingof Co(II) as a probe a Me(II)-S coordination occurring.

207

Changes to the Seventh Draft—Wednesday, September 26th Version

1. To construct and produce a family of ß-lactamases, as model proteins, using sitedirected mutagenesis and gene fusion techniques. This model proteins will allow asystematic characterization of IMAC for protein purification and as an analyticaltechnique for identification and characterization of metalloproteins.

1. 2 . To demonstrate that the portion of elution profile which are without anycontaminant proteins could be utilized for one step purification of target proteinfrom a cell extract by directing the elution of a target protein towards those areaemploying gene fusion techniques.

2. 3 . To assess whether a protein demonstrating multiple binding site for a metal ionwill utilize one site preferentially or multiple attachment will occur; and examinethe implications of these results for protein purification strategies whenemploying metal binding affinity peptides.

3. 4 . To examine a possible role for IMAC as a simple and rapid tool for identification ofzinc and other metal binding proteins in different preparations.

4. 5 . Contrast engineered ß-lactamases ß -lactamases (i.e. contain same terminalsequence but difference amounts of background histidine have been removed) towild type with DSC Differential Scanning Calorimetry (DSC) to ensure theengineered protein is not conformationally heterogeneous and to characterize anyalteration of conformation, or variation in crosslinking pattern, that may ofoccurred.

6. Contrast metal binding behavior of engineered proteins and native protein withDSC , IMAC, CD spectroscopy, and binding studies. Determine, for example, if thepresence of the added fragment can suppress the effects the metal ions have on theenzyme’s endotherm due to the fragment being able to bind the metal ionpreferentially.

208

Changes to the Eighth Draft—Thursday, September 27th Version

1. To construct and produce a family of ß ß- lactamases, as model proteins, using sitedirected mutagenesis and gene fusion techniques. This model proteins will allow asystematic characterization of IMAC for protein purification and as an analyticaltechnique for identification and characterization of metalloproteins.

2. To demonstrate that the portion of elution profile which are without anycontaminant proteins could be utilized for one step purification of target proteinfrom a cell extract by directing the elution of a target protein towards those areaemploying gene fusion techniques.

3. To assess whether a protein demonstrating multiple binding site for a metal ionwill utilize one site preferentially or multiple attachment will occur; and examinethe implications of these results for protein purification strategies whenemploying metal binding affinity peptides.

4. To examine a possible role for IMAC as a simple and rapid tool for identification ofzinc and other metal binding proteins in different preparations.

5. Contrast engineered ß ß- lactamases (i.e. contain same terminal sequence butdifference amounts of background histidine have been removed) to wild type withDifferential Scanning Calorimetry (DSC) to ensure the engineered protein is notconformationally heterogeneous and to characterize any alteration ofconformation, or variation in crosslinking pattern, that may of occurred. occurr .

6. Contrast metal binding behavior of engineered proteins and native protein withDSC, IMAC, CD spectroscopy, and binding studies. Determine, for example, if thepresence of the added fragment can suppress the effects the metal ions have on theenzyme’s endotherm due to the fragment being able to bind the metal ionpreferentially.

209

Changes to the Ninth Draft—Friday, September 28th Version

1. To construct and produce a family of ß-lactamases, as model proteins, using sitedirected mutagenesis and gene fusion techniques. This model proteins will allow asystematic characterization of IMAC for protein purification and as an analyticaltechnique for identification and characterization of metalloproteins.

2. To demonstrate that the portion of elution profile which are without anycontaminant contaminating proteins could be utilized for one step purification oftarget protein from a cell extract by directing the elution of a target proteintowards those area employing through the application of gene fusion techniques.

3. To assess whether a protein demonstrating multiple binding site for a metal ionwill utilize one site preferentially or multiple attachment will occur; and examinethe implications of these results for protein purification strategies whenemploying metal binding affinity peptides.

4. To examine a possible role for IMAC as a simple and rapid tool for identification ofzinc and other metal binding proteins in different preparations.

5. Contrast engineered ß-lactamases (i.e. contain same terminal sequence butdifference amounts of background histidine have been removed) to wild type withDifferential Scanning Calorimetry (DSC) to ensure the engineered protein is notconformationally heterogeneous and to characterize any alteration ofconformation, or variation in crosslinking pattern, that may occurr. occur .

6. Contrast metal binding behavior of engineered proteins and native protein withDSC, IMAC, CD spectroscopy, and binding studies. Determine, for example, if thepresence of the added fragment can suppress the effects the metal ions have on theenzyme’s endotherm due to the fragment being able to bind the metal ionpreferentially.

210

Changes to the Tenth Draft—Saturday, September 29th First Version

1. To Construct and produce a family of ß-lactamases, as model proteins, using sitedirected mutagenesis and gene fusion techniques. This These model proteins willallow for a systematic characterization of IMAC for protein purification and as ananalytical technique for the identification and characterization ofmetalloproteins.

2. Determine how sample pretreatment (e.g. remove bound metal ions first) may alterthe elution profile of host cell proteins so that further information on the relativeabundance of zinc binding and nonbinding proteins is obtained. Also ascertainwhether alternative “window” opportunites exist via different pretreatments.

2. 3. To Demonstrate that the portion portions of elution profile which are without anythat are deviod of contaminating proteins could be utilized for an one-steppurification of a target protein from a cell extract by directing the elution of atarget protein towards those area portions through the application of gene fusiontechniques.

3. 4 . To Assess whether a protein demonstrating multiple binding site for a metal ionwill utilize one site preferentially or multiple attachment will occur; and examinethe implications of these results for protein purification strategies whenemploying metal binding affinity peptides.

5. Contrast engineered ß-lactamases (i.e. contain same terminal sequence butdifference amounts of background histidine have been removed) to wild type withDifferential Scanning Calorimetry (DSC) to ensure the engineered protein is notconformationally heterogeneous and to characterize any alteration ofconformation, or variation in crosslinking pattern, that may occur.

6. 5 . Contrast Characterize the model proteins and then contrast the metal bindingbehavior of engineered proteins and native protein with DSC , IMAC, lightspectroscopy, CD spectroscopy, and metal binding studies. Determine, for example,if the presence of the added fragment a tail can suppress the effects the metal ionshave on the effect metal ion binding has on an enzyme’s endotherm due to the fragment tai l being able to bind the metal ion preferentially. Then relate thismolecular-level information to the IMAC results.

211

Changes to the Eleventh Draft—Saturday, Sept. 29th Second Version

1. Construct and produce a family of ß-lactamases, as model proteins, using sitedirected mutagenesis and gene fusion techniques. These model proteins will allowfor a systematic characterization of IMAC for protein purification and as ananalytical technique for the identification and characterization ofmetalloproteins.

2. Determine how sample pretreatment (e.g. remove bound metal ions first) may alterthe elution profile of host cell proteins so that further information on the relativeabundance of zinc binding and nonbinding proteins is obtained. Also ascertainwhether alternative “window” opportunites exist via different pretreatments.

3. Demonstrate that the portions of elution profile that are deviod of contaminatingproteins could be utilized for an one-step purification of a target protein from a cellextract by directing the elution of a target protein towards those portions throughthe application of gene fusion techniques.

4. Assess whether a protein demonstrating multiple binding site for a metal ion willutilize one site preferentially or multiple attachment will occur; and examine theimplications of these results for protein purification strategies when employingmetal binding affinity peptides.

5. Characterize the model proteins and then contrast the metal binding behavior ofengineered proteins and native protein with DSC, light spectroscopy, CDspectroscopy, and metal binding studies. Determine, for example, if the presence ofa tail can suppress the effect metal ion binding has on an enzyme’s endotherm dueto the tail being able to bind the metal ion preferentially. Then relate thismolecular-level information to the IMAC results.

6. Apart from the aim of using IMAC as a one-step protein isolation procedure,examine the results from the standpoint of using IMAC as a simple and rapid toolfor identifying zinc and other metal binding proteins in different preparations.

212

Changes to the Twelfth Draft—Sunday, September 30th Version

1. Construct and produce a family of ß-lactamases, as model proteins, using sitedirected mutagenesis and gene fusion techniques. These model proteins will allowfor a systematic characterization of IMAC for protein purification and as ananalytical technique for the identification and characterization ofmetalloproteins.

2. Determine how sample pretreatment (e.g. remove bound metal ions first) may alterthe elution profile of host cell proteins so that further information on the relativeabundance of zinc binding and nonbinding proteins is obtained. Also ascertainwhether alternative “window” opportunites exist via different pretreatments.

3. Demonstrate that the portions of elution profile that are deviod devoid ofcontaminating proteins could be utilized for an one-step purification of a targetprotein from a cell extract by directing the elution of a target protein towards thoseportions through the application of gene fusion techniques.

4. Assess whether a protein demonstrating multiple binding site for a metal ion willutilize one site preferentially or multiple attachment will occur; and examine theimplications of these results for protein purification strategies when employingmetal binding affinity peptides.

5. Characterize the model proteins and then contrast the metal binding behavior ofengineered proteins and native protein with DSC, light spectroscopy, CDspectroscopy, and metal binding studies. Determine, for example, if the presence ofa tail can suppress the effect metal ion binding has on an enzyme’s endotherm dueto the tail being able to bind the metal ion preferentially. Then relate thismolecular-level information to the IMAC results.

6. Apart from the aim of using IMAC as a one-step protein isolation procedure,examine the results from the standpoint of using IMAC as a simple and rapid toolfor identifying zinc and other metal binding proteins in different preparations.

213

Changes to the Thirteenth & Final Draft—Monday, October 1st Version

1. Construct and produce a family of ß-lactamases, as model proteins, using sitedirected mutagenesis and gene fusion techniques. These model proteins will allowfor a systematic characterization of IMAC for protein purification and as ananalytical technique for the identification and characterization ofmetalloproteins.

2. Determine how sample pretreatment (e.g. remove bound metal ions first) may alterthe elution profile of host cell proteins so that further information on the relativeabundance of zinc binding and nonbinding proteins is obtained. Also ascertainwhether alternative “window” opportunities exist via can be generated by using different pretreatments.

3. Demonstrate that the portions of elution profile that are devoid of contaminatingproteins could be utilized for an one-step purification of a target protein from a cellextract by directing the elution of a target protein towards those portions throughthe application of gene fusion techniques.

4. Assess whether a protein demonstrating multiple binding site for a metal ion willutilize one site preferentially or multiple attachment will occur; and examine theimplications of these results for protein purification strategies when employingmetal binding affinity peptides.

5. Characterize the model proteins and then contrast the metal binding behavior ofengineered proteins and native protein with DSC, light spectroscopy, CDspectroscopy, and metal binding studies. Determine, for example, if the presence ofa tail can suppress the effect metal ion binding has on an enzyme’s endotherm dueto the tail being able to bind the metal ion preferentially. Then relate thismolecular-level information to the IMAC results.

6. Apart from the aim of using IMAC as a one-step protein isolation procedure,examine the results from the standpoint of using IMAC as a simple and rapid toolfor identifying zinc and other metal binding proteins in different preparations.

214

The Final Draft—Monday, October 1st Version

1. Construct and produce a family of ß-lactamases, as model proteins, using sitedirected mutagenesis and gene fusion techniques. These model proteins will allowfor a systematic characterization of IMAC for protein purification and as ananalytical technique for the identification and characterization ofmetalloproteins.

2. Determine how sample pretreatment (e.g. remove bound metal ions first) may alterthe elution profile of host cell proteins so that further information on the relativeabundance of zinc binding and nonbinding proteins is obtained. Also ascertainwhether alternative “window” opportunities can be generated by using differentpretreatments.

3. Demonstrate that the portions of elution profile that are devoid of contaminatingproteins could be utilized for an one-step purification of a target protein from a cellextract by directing the elution of a target protein towards those portions throughthe application of gene fusion techniques.

4. Assess whether a protein demonstrating multiple binding site for a metal ion willutilize one site preferentially or multiple attachment will occur; and examine theimplications of these results for protein purification strategies when employingmetal binding affinity peptides.

5. Characterize the model proteins and then contrast the metal binding behavior ofengineered proteins and native protein with DSC, light spectroscopy, CDspectroscopy, and metal binding studies. Determine, for example, if the presence ofa tail can suppress the effect metal ion binding has on an enzyme’s endotherm dueto the tail being able to bind the metal ion preferentially. Then relate thismolecular-level information to the IMAC results.

6. Apart from the aim of using IMAC as a one-step protein isolation procedure,examine the results from the standpoint of using IMAC as a simple and rapid toolfor identifying zinc and other metal binding proteins in different preparations.

215

Appendix J—Evolution of the Proposal’s Three Major Sections

The Table 1 is organized as follows. II is the percentage of the proposal taken up bythe Significance section over the 14 drafts; a is the IMAC Background and LiteratureSurvey subsection; b is the DSC Background and Literature Survey subsection, and; c is theContributions of the Proposed Research subsection:

1 2 3 4 5 6 7 8 9 10 11 12 13 14

II 25 4 4 4 19 19 15 22 22 20 21 21 22 20

a (13) (13) (10) (9) (9) (8) (8) (8) (8) (8)

b (10) (10) (9) (9) (9) (7) (6)

c (6) (6) (5) (3) (3) (3) (4) (4) (7) (6)

Table 1: The development and percentage breakdown of the Significance section of theproposal.

Table 2 is organized as follows. III is the percentage of the proposal taken up bythe Preliminary Efforts section over the 14 drafts; a is the Zn(II)-IDA IMAC Experimentssubsection; b is the Cu(II)-IDA IMAC Experiments subsection; c is the Discussion of theIMAC Results subsection; d is the Overview of Differential Scanning CalorimetryExperiments subsection; e is the Differential Scanning Calorimetry Results subsection; f isthe Discussion of Differential Calorimetry Results subsection, and; g is the Summary andAnticipated Utility of Differential Scanning Calorimetry Studies subsection:

216

1 2 3 4 5 6 7 8 9 10 11 12 13 14

III 8 84 76 74 50 50 40 32 33 24 24 22 21 22

a (7) (8) (6) (5) (5) (5) (5) (4) (4) (4)

b (5) (4) (3) (3) (2) (2) (1) (1) (1)

c (15) (9) (8) (6) (7) (6) (6) (6) (5) (6)

d (16) (8) (7) (3) (3) (2) (2) (2) (2) (2) (2) (2) (2)

e (8) (7) (3) (3) (2) (2) (2) (2) (2) (2) (2) (2)

f (37) (36) (3) (13) (10) (8) (8) (5) (5) (5) (5) (5)

g (23) (24) (9) (9) (8) (6) (6) (2) (2) (2) (2) (2)

Table 2: The development and percentage breakdown of the Preliminary Efforts section ofthe proposal.

Table 3 is organized as follows. IV is the percentage of the proposal taken up bythe Research Plan section over the 14 drafts; a is the Choice and Construction of ModelProteins subsection; b is the Characterization of Model Proteins subsection; c is the IMACExperiments subsection; d is the Metal Binding Studies subsection, and; e is theCollaborative Arrangements and Time Table subsection:

1 2 3 4 5 6 7 8 9 10 11 12 13 14

V 32 20 20 27 29 30 27 27 28 28 30

a (32) (10) (10) (8) (8) (10) (9) (9) (9) (9) (11)

b (10) (10) (6) (4) (4) (2) (2) (2) (2) (2)

c (4) (4) (5) (4) (4)

d (13) (17) (16) (12) (12) (12) (13) (12)

e (1)

Table 3: The development and percentage breakdown of the Research Plan section of theproposal.

217

218

Bibliography

Allen, W. (1966, 1989). Getting Even. Without Feathers, Getting Even, Side Effects. NY,

NY: Quality Paperback.

Anderson, P. V., Brockmann, J. R., & Miller, C. R. (Eds.). (1983). New Essays in Technical

and Scientific Communication. NY, NY: Baywood P.

Anson, C. M. & Forsberg, L. L. (1990). Moving Beyond the Academic Community:

Transitional Stages in Professional Writing. Written Communication, 7 (2), 200-

231.

Applebee, A. N. (1986). Problems in Process Approaches: Toward a Reconceptualization of

Process Instruction. The Teaching of Writing: Eighty-fifth Yearbook of the

National Society for the Study of Education, Part II. A. R. Petrosky & D.

Bartholomae (Eds.). Chicago, IL: U of Chicago P, 95-113.

Atlas, M. (1979). Addressing an Audience: A Study of Expert-Novice Differences in

Writing. Document Design Project Technical Report No. 3. Pittsburgh, PA:

Carnegie Mellon, Communications Design Center.

Barnes, B. (1977). Interests and the Growth of Knowledge. London, England: Routledge and

Kegan Paul.

Barnum, C. & Fischer, R. (1984). Engineering Technologists as Writers: Results of a Survey.

Technical Communication, 31 (2), 9-11.

219

Bartholomae, D. (1985). Inventing the University. When a Writer Can’t Write: Studies in

Writer’s Block and Other Composing-Process Problems. M. Rose (Ed.). NY, NY:

Guilford P, 134-165.

Bassok, M., & Holyoak, K. J. (1985). Schema-based Interdomain Transfer Between

Isomorphic Algebra and Physics Problems. Manuscript in Preparation. Ann Arbor,

MI: U of Michigan.

Bazerman, C. (1984). The Writing of Scientific Non-fiction: Contexts, Choices, Constraints.

Pre/Text, 5, 39-74.

Bazerman, C. (1988). Shaping Written Knowledge: The Genre and Activity of the

Experimental Article in Science. Madison, WI: The U of Wisconsin P.

Bazerman, C. (1991). How Natural Philosophers Can Cooperate: The Literary Technology

of Coordinated Investigation in Joseph Priestley’s “History and Present State of

Electricity” (1767). Textual Dynamics of the Professions: Historical and

Contemporary Studies of Writing in Professional Communities. C. Bazerman & J.

Paradis (Eds.). Madison, WI: U of Wisconsin P, 13-44.

Bereiter, C. & Scardamalia, M. (1987). The Psychology of Written Composition.

Hillsdale, NJ: Lawrence Erlbaum.

Berelson, B. (1971). Content Analysis in Communication Research. NY, NY: Free P.

Berkenkotter, C., Huckin, T. N., & Ackerman, J. (1991). Social Context and Socially

Constructed Texts: The Initiation of a Graduate Student into a Writing Research

Community. Textual Dynamics of the Professions: Historical and Contemporary

220

Studies of Writing in Professional Communities. C. Bazerman & J. Paradis (Eds.).

Madison, WI: U of Wisconsin P, 191-215.

Berlin, J. A. (1987). Rhetoric and Reality: Writing Instruction in American Colleges, 1900-

1985. Carbondale, IL: Southern Illinois UP.

Berlin, J. A. (1988). Rhetoric and Ideology in the Writing Class. College English, 50, 477-

494.

Berlin, J. A. & Inkster, R. P. (1980). Current-Traditional Rhetoric: Paradigm and Practice.

Freshman English News, 8 (3), 1-18.

Bernhardt, S. A. (1985). The Writer, the Reader, and the Scientific Text. Journal of

Technical Writing and Communication, 15 (2), 163-174.

Bhaskar, R. & Simon, H. A. (1977). Problem Solving in Semantically Rich Domains: An

Example from Engineering Thermodynamics. Cognitive Science, 1, 193-215.

Bizzell, P. (1982a). Cognition, Convention, and Certainty: What We Need to Know about

Writing. Pre/Text, 3, 213-244.

Bizzell, P. (1982b). College Composition: Initiation into the Academic Discourse

Community. Curriculum Inquiry, 12 (2), 191-207.

Bizzell, P. (1988). Cultural Criticism: A Social Approach to Writing. Paper presented at

the Conference on College Composition and Communication.

Bjerknes, G., Ehn, P., & Kyng, M. (1987). Computers and Democracy—A Scandinavian

Challenge . Aldershot, England: Avebury.

221

Bloor, D. (1976). Knowledge and Social Imagery. London, England: Routledge and Kegan

Paul.

Blyler, N. R. (1989). Purpose and Professional Writers. The Technical Writing Teacher, 16

(1), 52-67.

Bradshaw, G., Langley, P., & Simon, H. A. (1983). Studying Scientific Discovery by

Computer Simulation. Science, 222, 971-975.

Brandt, D. (1986). Toward an Understanding of Context in Composition. Written

Communication, 3 (2), 139-157.

Brenner, M., Brown, J., & Canter, D. (Eds.). (1985). The Research Interview: Uses and

Approaches . NY, NY: Academic P.

Britton, J. (1978). The Composing Processes and the Functions of Writing. Research on

Composing: Points of Departure. C. R. Cooper & L. Odell (Eds.). Urbana, IL:

National Council of Teachers of English, 13-28.

Britton, B. K. & Black, J. B. (Eds.). (1985). Understanding Expository Text: A Theoretical

and Practical Handbook for Analyzing Explanatory Text. Hillsdale, NJ: Lawrence

Erlbaum.

Broadhead, G. J. & Freed, R. C. (1986). The Variables of Composition: Process and Product

in a Business Setting. Carbondale, IL: Southern Illinois UP.

Brown, J. & Canter, D. (1985). The Uses of Explanation in the Research Interview. The

Research Interview: Uses and Approaches. M. Brenner, J. Brown, & D. Canter

(Eds.). NY, NY: Academic P, 217-245.

222

Brown, J. S., Collins, A., & Duguid, P. (1989). Situated Cognition and the Culture of

Learning. Educational Researcher, 18 (1), 32-42.

Brown, Jr., R. L. & Herndl, C. G. (1986). An Ethnographic Study of Corporate Writing: Job

status as Reflected in Written Text. Functional Approaches to Writing: Research

Perspectives. B. Couture (Ed.). Norwood, NJ: Ablex, 11-28.

Bruffee, K. A. (1984). Collaborative Learning and the Conversation of Mankind. College

English, 46 (7), 635-652.

Bruffee, K. A. (1986). Social Construction, Language, and the Authority of Knowledge: A

Bibliographic Essay. College English, 48 (8), 773-787.

Budish, B. E. & Sandhusen, R. L. (1989). The Short Proposal: Versatile Tool for

Communicating Corporate Culture in Competitive Climates. IEEE Transactions on

Professional Communication, 32 (2), 81-85.

Burke, K. (1941, 1973). The Philosophy of Literary Form: Studies in Symbolic Action, 3rd

Edition. Berkeley, CA: U of California P.

Burke, K. (1945). A Grammar of Motives. NY, NY: Prentice-Hall.

Burke, K. (1978). Rhetoric, Poetics, and Philosophy. Rhetoric, Philosophy, and

Literature: An Exploration. D. Burks (Ed.). West Lafayette, IN: Purdue UP, 15-34.

Burtis, P. J., Bereiter, C., Scardamalia, M., & Tetroe, J. (1983). The Development of

Planning in Writing. Explorations in the Development of Writing. B. M. Kroll & G.

Wells (Eds.). NY, NY: Wiley, 153-174.

223

Campbell, P. N. (1973). Poetic-rhetorical, Philosophical, and Scientific Discourse.

Philosophy and Rhetoric, 6, 1-29.

Campbell, P. N. (1975). The “Personae” of Scientific Discourse. Quarterly Journal of

Speech, 61, 391-405.

Canter, D., Brown, J., & Groat, L. (1985). A Multiple Sorting Procedure for Studying

Conceptual Systems. The Research Interview: Uses and Approaches. M. Brenner, J.

Brown, & D. Canter (Eds.). NY, NY: Academic P, 79-114.

Card, S., Moran, T., & Newell, A. (1983). The Psychology of Human-Computer Interaction.

Hillsdale, NJ: Lawrence Erlbaum.

Carter, M. (1990). The Idea of Expertise: An Exploration of Cognitive and Social

Dimensions of Writing. College Composition and Communication, 41 (3), 265-286.

Charney, D. (1987). Comprehending Non-Linear Text: The Role of Discourse Cues and

Reading Strategies. Hypertext ’87 Papers. Chapel Hill, NC: U of North Carolina,

109-120.

Chi, M. T. H., Glaser, R., & Rees, E. (1982). Expertise in Problem Solving. Advances in the

Psychology of Human Intelligence, Vol. 1. R. J. Sternberg (Ed.). Hillsdale, NJ:

Lawrence Erlbaum, 7-75.

Chubin, D. E. & Restivo, S. (1983). The “Mooting” of Science Studies: Research Programmes

and Science Policy. Science Observed. K. D. Knorr-Cetina & M. Mulkay (Eds.).

London, England: Sage, 53-83.

224

Clifford, J. (1980). Fieldwork, Reciprocity, and the Making of Ethnographic Texts: The

Example of Maurice Leenhardt. Man, 15, 518-532.

Clifford, J. (1982). Person and Myth: Maurice Leenhardt in the Melanesian World.

Berkeley, CA: U of California P.

Clifford, J. & Marcus, G. E. (Eds.). (1986). Writing Culture. Berkeley, CA: U of California

P.

Coe, R. M. (1987). An Apology for Form; or, Who Took the Form Out of the Process? College

English, 49 (1), 13-28.

Cole, S., Rubin, L., & Cole, J. R. (1977). Peer Review and the Support of Science Scientific

American, 237, 34-41.

Collins, H. M. (1981a). Son of the Seven Sexes: The Social Destruction of a Physical

Phenomenon. Social Studies of Science, 11 (1), 33-62.

Collins, H. M. (1981b). Stages in the Empirical Programme of Relativism. Social Studies

of Science, 11 (1), 3-10.

Collins, H. M. (1982). Knowledge, Norms and Rules in the Sociology of Science. Social

Studies of Science, 12 (2), 299-309.

Cooper, M. & Holzman, M. (1983). Talking About Protocols. College Composition and

Communication, 34 (3), 284-296.

Cooper, C. R. & Odell, L. (1976). Considerations of Sound in the Composing Process of

Published Writers. Research in the Teaching of English, 10 (1), 103-115.

225

Couture, B. (1986). Functional Approaches to Writing: Research Perspectives. Norwood,

NJ: Ablex.

Crowley, S. (1985). Invention in Nineteenth-Century Rhetoric. College Composition and

Communication, 36 (1), 51-60.

Culler, D. (1968). The Darwinian Revolution and Literary Form. The Art of Victorian

Prose. G. Levine & W. Madden (Eds.). NY: NY: Oxford UP, 224-246.

De Bakey, L. (1976). The Persuasive Proposal. Journal of Technical Writing and

Communication, 6 (1), 5-25.

Debs, M. B. (1989). Collaborative Writing in Industry. Technical Writing: Theory and

Practice. B. E. Fearing & W. K. Sparrow (Eds.). NY, NY: The Modern Language

Association of America, 33-42.

Doheny-Farina, S. (1986). Writing in an Emerging Organization. Written Communication,

3 (2) , 158-185.

Doheny-Farina, S. (1989). A Case Study of One Adult Writing in Academic and

Nonacademic Discourse Communities. Worlds of Writing: Teaching and Learning in

Discourse Communities of Work. C. B. Matalene (Ed.). NY, NY: Random House, 17-

42.

Doheny-Farina, S. (1991). Creating a Text/Creating a Company: The Role of a Text in the

Rise and Decline of a New Organization. Textual Dynamics of the Professions:

Historical and Contemporary Studies of Writing in Professional Communities. C.

Bazerman & J. Paradis (Eds.). Madison, WI: U of Wisconsin P, 306-335.

226

Doheny-Farina, S. & Odell, L. (1985). Ethnographic Research on Writing: Assumptions

and Methodology. Writing in Nonacademic Settings. L. Odell & D. Goswami

(Eds.). London, England: The Guilford P, 503-534.

Duffy, T. M., Higgins, L., Mehlenbacher, B., Cochran, C., Wallace, D., Hill, C. A., Haugen,

D., McCaffrey, M., Burnett, R., Sloane, S., & Smith, S. (1989). Models for the

Design of Instructional Text. Reading Research Quarterly, 24 (4), 434-456.

Duffy, T. M. & Kabance, P. (1982). Testing a Readable Writing Approach to Text Revision.

Journal of Educational Psychology, 74, 733-748.

Duffy, T. M., Mehlenbacher, B., & Palmer, J. E. (1989). The Evaluation of Online Help

Systems: A Conceptual Model. The Society of Text: Hypertext, Hypermedia, and

the Social Construction of Information. E. Barrett (Ed.). Cambridge, MA: MIT P,

362-387.

Duffy, T. M., Palmer, J. E., & Mehlenbacher, B. (1992). Online Help: Design and

Evaluation. Norwood, NJ: Ablex.

Duffy, T. M. & Waller, R. (Eds.). (1985). Designing Usable Texts. Orlando, FL: Academic

P.

Duit, R. (1991). Students’ Conceptual Frameworks: Consequences for Learning Science. The

Psychology of Learning Science. S. M. Glynn, R. H. Yeany, & B. K. Britton (Eds.).

Hillsdale, NJ: Lawrence Erlbaum, 65-85.

227

Dycus, R. D. (1977). The Effect of Proposal Appearance on the Technical Evaluation

Scoring of Government Proposals. Journal of Technical Writing and Communication,

7 (4) , 285-295.

Eaves, G. N. (1984). Preparation of the Reseach-Grant Application: Opportunities and

Pitfalls. Grants Magazine, 7 (3), 151-157.

Ede, L. & Lunsford, A. (1984). Audience Addressed/Audience Invoked: The Role of

Audience in Composition Theory and Pedagogy. College Composition and

Communication, 35 (2), 155-171.

Ericsson, K. A. & Simon, H. A. (1980). Verbal Reports as Data. Psychological Review, 87,

215-251.

Ericsson, K. A. & Simon, H. A. (1984). Protocol Analysis: Verbal Reports as Data.

Cambridge, MA: MIT P.

Fahnestock, J. (1986). Accommodating Science: The Rhetorical Life of Scientific Facts.

Written Communication, 3 (3), 275-296.

Faigley, L. (1985). Nonacademic Writing: The Social Perspective. Writing in

Nonacademic Settings. L. Odell & D. Goswami (Eds.). London, England: The

Guilford P, 231-248.

Faigley, L. & Hansen, K. (1985). Learning to Write in the Social Sciences. College

Composition and Communication, 36 (2), 140-149.

Fearing, B. E. & Sparrow, W. K. (Eds.). (1989). Technical Writing: Theory and Practice.

NY, NY: The Modern Language Association of America.

228

Feyerabend, P. (1975). Against Method: Outline of an Anarchistic Theory of Knowledge.

London, England: New Left Books.

Fisher, W. R. (1984). Narration as a Human Communication Paradigm: The Case of Public

Moral Argument. Communication Monograph, 51, 1-22.

Fisher, W. R. (1987). Human Communication as Narration: Toward a Philosophy of

Reason, Value, and Action. Columbia, SC: U of South Carolina P.

Flower, L. (1989). Cognition, Context, and Theory Building. College Composition and

Communication, 40 (3), 282-311.

Flower, L., Schriver, K. A., Carey, L., Haas, C., & Hayes, J. R. (1989). Planning in Writing:

The Cognition of a Constructive Process. Center for the Study of Writing Technical

Report No. 34. Berkeley, CA & Pittsburgh, PA: U of California at Berkeley and

Carnegie Mellon.

Floyd, C. (1987). Outline of a Paradigm Change in Software Engineering. Computers and

Democracy—A Scandinavian Challenge. G. Bjerknes, P. Ehn., & M. Kyng (Eds.).

Aldershot, England: Avenbury, 191-210.

Freed, R. C. (1987). A Meditation on Proposals and Their Backgrounds. Journal of

Technical Writing and Communication, 17 (2), 157-163.

Freed, R. C. & Roberts, D. D. (1989). The Nature, Classification, and Generic Structure of

Proposals. Journal of Technical Writing and Communication, 19 (4), 317-351.

Garfinkel, H. (1967). Studies in Ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall.

229

Geertz, C. (1973). The Interpretation of Cultures. NY, NY: Basic Books.

Geertz, C. (1983). Local Knowledge: Further Essays in Interpretive Anthropology. NY,

NY: Basic Books.

George, A. L. (1959). Quantiative and Qualitative Approaches to Content Analysis. NY,

NY: Rand.

Gilbert, G. N. (1977). Referencing as Persuasion. Social Studies of Science, 7 (1), 113-122.

Gilbert, G. N. & Mulkay, M. (1980). Contexts of Scientific Discourse: Social Accounting in

Experimental Papers. The Social Process of Scientific Investigation, Sociology of

the Sciences Yearbook, Vol. 4. K. Knorr, R., Krohn, & R. Whitley (Eds.).

Dordrecht: D. Reidel, 269-294.

Gilbert, G. N. & Mulkay, M. (1984). Opening Pandora’s Box: A Sociological Analysis of

Scientists’ Discourse. Cambridge, England: Cambridge UP.

Gilsdorf, J. W. (1986). Executives’ and Academics’ Perceptions on the Need for Instruction

in Written Persuasion. The Journal of Business Communication, 23 (4), 55-68.

Glynn, S. M. (1991). Explaining Science Concepts: A Teaching-with-Analogies Model. The

Psychology of Learning Science. S. M. Glynn, R. H. Yeany, & B. K. Britton (Eds.).

Hillsdale, NJ: Lawrence Erlbaum, 219-239.

Glynn, S. M., Yeany, R. H., & Britton, B. K. (Eds.). (1991). The Psychology of Learning

Science. Hillsdale, NJ: Lawrence Erlbaum.

230

Gökalp, I. (1990). Turbulent Reactions: Impact of New Instrumentation on a Borderland

Scientific Domain. Science, Technology, & Human Values, 15 (3), 284-304.

Goldberg, J. (1988). Anatomy of a Scientific Discovery: The Race to Discover the Secret of

Human Pain and Pleasure. NY, NY: Bantam Books.

Gragson, G. & Selzer, J. (1990). Fictionalizing the Readers of Scholarly Articles in

Biology. Written Communication, 7 (1), 25-58.

Grammatik Mac. (1989). Reference Software.

Greeno, J. G. (1988). Situations, Mental Models, and Generative Knowledge. IRL Report 88-

0005, Palo Alto, CA: Institute for Research on Learning.

Gregg, L. & Steinberg, E. R. (Eds.). (1980). Cognitive Processes in Writing: An

Interdisciplinary Approach. Hillsdale, NJ: Lawrence Erlbaum.

Gross, A. G. (1985). The Form of the Experimental Paper: A Realization of the Myth of

Induction. Journal of Technical Writing and Communication, 15 (1), 15-26.

Gross, A. G. (1989). The Rhetorical Invention of Scientific Invention: The Emergence and

Transformation of a Social Norm. Rhetoric in the Human Sciences. H. W. Simons

(Ed.). London, England: Sage, 89-107.

Gross, A. G. (1990a). Discourse on Method: The Rhetorical Analysis of Scientific Texts.

Pre/Text, 9 (3-4), 169-185.

Gross, A. G. (1990b). The Rhetoric of Science. Cambridge, MA: Harvard UP.

231

Gusfield, J. (1976). The Literary Rhetoric of Science: Comedy and Pathos in Drinking

Driver Research. American Sociological Review, 41, 16-34.

Gustafson, T. (1975). The Controversy Over Peer Review. Science, 190, 1060-1066.

Hairston, M. (1982). The Winds of Change: Thomas Kuhn and the Revolution in the

Teaching of Writing. College Composition and Communication, 33 (1), 76-88.

Halliday, M. A. K. (1978). Language as Social Semiotic: The Social Interpretation of

Language and Meaning. Baltimore, MD: University Park P.

Halliday, M. A. K. & Hasan, R. (1976). Cohesion in English. London, England: Longman.

Halloran, S. M. & Bradford, A. (1984). Figures of Speech in the Rhetoric of Science and

Technology. Essays on Classical Rhetoric and Modern Discourse. R. J. Connors, L. S.

Ede, & A. A. Lunsford (Eds.). Carbondale, IL: Southern Illinois UP, 179-192.

Hamilton, D. (1978). Writing Science. College English, 40 (1), 32-40.

Hannay, N. B. & McGinn, R. E. (1980). The Anatomy of Modern Technology: Prolegomenon

to an Improved Public Policy for the Social Management of Technology. Daedalus,

109 (1) , 25-53.

Harmon, J. E. (1989). Development of the Modern Technical Article. Technical

Communication, 36 (1), 33-38.

Harvey, B. (1981). Plausibility and the Evaluation of Knowledge: A Case Study of

Experimental Quantum Mechanics. Social Studies of Science, 11(1), 95-130.

232

Hill, C. A., Wallace, D. L., & Haas, C. (in press). Revising Online: Computer Technologies

and the Revising Process. Computers and Composition.

Harrison, T. M. (1987). Frameworks for the Study of Writing in Organizational Contexts.

Written Communication, 4 (1), 3-23.

Haselkorn, M. P. (1985). Proposals. Research in Technical Communication: A

Bibliographic Sourcebook. M. G. Moran & D. Journet (Eds.). Westport, CN:

Greenwood, 255-283.

Hayes, J. R. (1989). Writing Research: The Analysis of a Very Complex Task. Complex

Information Processing: The Impact of Herbert A. Simon. D. Klahr & K. Kotovsky

(Eds.). Hillsdale, NJ: Lawrence Erlbaum, 209-234.

Hayes, J. R. & Flower, L. (1980). Identifying the Organization of Writing Processes.

Cognitive Processes in Writing: An Interdisciplinary Approach. L. Gregg & E.

Steinberg (Eds.). Hillsdale, NJ: Lawrence Erlbaum, 3-30.

Hayes, J. R., Flower, L., Schriver, K. A., Stratman, J. F., & Carey, L. (1987). Cognitive

Processes in Revision. Advances in Applied Psycholinguistics, Vol. 2: Reading,

Writing, and Language Learning. S. Rosenberg (Ed.). Cambridge, England:

Cambridge UP, 176-240.

Heath, S. B. & Branscombe, A. (1985). Intelligent Writing in an Audience Community:

Teacher, Students, and Researchers. The Acquisition of Written Language. S. W.

Freedman (Ed.). Norwood, NJ: Ablex, 3-32.

233

Herndl, C. G., Fennell, B. A., & Miller, C. R. (1991). Understanding Failures in

Organizational Discourse: The Accident at Three Mile Island and the Shuttle

Challenger Disaster. Textual Dynamics of the Professions: Historical and

Contemporary Studies of Writing in Professional Communities. C. Bazerman & J.

Paradis (Eds.). Madison, WI: U of Wisconsin P, 279-305.

Herrington, A. (1985). Writing in Academic Settings: A Study of the Contexts for Writing

in Two Chemical Engineering Courses. Research in the Teaching of English, 19 (4),

331-361.

Janich, P. (1978). Physics—Natural Science or Technology? The Dynamics of Science and

Technology, Sociology of the Sciences Yearbook, Vol. 2. W. Krohn, E. Layton, & P.

Weingart (Eds.). Dordrecht: D. Reidel, 3-27.

Journet, D. (1990). Writing, Rhetoric, and the Social Construction of Scientific Knowledge.

IEEE Transactions on Professional Communication, 33 (4), 162-167.

Journet, D. (1991). Ecological Theories as Cultural Narratives: F. E. Clements’s and H. A.

Gleason’s “Stories” of Community Succession. Written Communication, 8 (4), 446-

472.

Kasperson, C. J. (1976). An Exploratory Analysis of Information Use by Innovative,

Productive, and Non-Productive Scientists and Engineers. Ph.D. Dissertation.

Troy, NY: Rensselaer.

Katz, S. B. (1992). Narration, Technical Communication, and Culture: “The Soul of a New

Machine” as Narrative Romance. Constructing Rhetorical Education: From the

234

Classroom to the Community. M. Secar & D. Charney (Eds.) Carbondale, IL:

Southern Illinois UP, 382-402.

Kaufer, D. S. & Geisler, C. (1989). Novelty in Academic Writing. Written Communication,

6 (3) , 286-311.

Kaufer, D. S., Hayes, J. R., & Flower, L. (1986). Composing Written Sentences. Research in

the Teaching of English, 20 (2), 121-140.

Kelso, J. A. (1980). Science and the Rhetoric of Reality. Central States Speech Journal, 31,

17-29.

Kennedy, G. E. (1983). Teaching Formal Proposals: A Versatile Minicourse in Technical

Writing. Journal of Technical Writing and Communication, 13 (2), 123-137.

Kenward, M. (1984). Peer Review and the Axe Murderers. New Scientist, 13.

Killingsworth, M. J. (1983). A Bibliography on Proposal Writing. IEEE Transactions on

Professional Communication, 26 (2), 79-83.

Kinneavy, J. L. (1983). Writing Across the Curriculum. Profession 83, 13-20.

Kintsch, W. & van Dijk, T. A. (1978). Towards a Model of Discourse Comprehension and

Production. Psychological Review, 85, 363-394.

Knorr-Cetina, K. D. & Mulkay, M. (1983). Emerging Principles in Social Studies of Science.

Science Observed. K. D. Knorr-Cetina & M. Mulkay (Eds.). London, England: Sage,

1-17.

235

Kotovsky, K., Hayes, J. R., & Simon, H. A. (1985). Why are Some Problems Hard?

Evidence from Tower of Hanoi. Cognitive Psychology, 17, 248-294.

Kuhn, T. S. (1970). The Structure of Scientific Revolutions. Chicago, IL: U of Chicago P.

Kulkarni, D. & Simon, H. A. (1988). The Processes of Scientific Discovery: The Strategy of

Experimentation. Cognitive Science, 12, 139-175.

Langley, P. (1981). Data-Driven Discovery of Physical Laws. Cognitive Science, 5, 31-54.

Langley, P., Simon, H. A., Bradshaw, G., & Zytkow, J. M. (1987). Scientific Discovery.

Cambridge, MA: MIT P.

Langley, P., Zytkow, J. M., Simon, H. A., & Bradshaw, G. (1983). Mechanisms for

Qualitative and Quantitative Discovery. Proceedings of the International

Machine Learning Workshop. Monticello, IL: U of Illinois, 121-132.

Larkin, J., McDermott, J., Simon, D. P., & Simon, H. A. (1980). Expert and Novice

Performance in Solving Physics Problems. Science, 208, 1335-1342.

Latour, B. (1988). A Relativistic Account of Einstein’s Relativity. Social Studies of

Science, 18 (1), 3-44.

Latour, B. & Woolgar, S. (1979). Laboratory Life: The Social Construction of Scientific

Facts. London, England: Sage.

Law, J. & Williams, R. (1982). Putting Facts Together: A Study of Scientific Persuasion.

Social studies of science, 12 (4), 535-558.

236

Lawrence, S. V. (1978). Grants: Fuel That Feeds Research. The Bulletin of the American

College of Physicians, 19, 18-26.

LeFevere, J. A. & Dixon, P. (1986). Do Written Instructions Need Examples? Cognition and

Instruction, 3 (1), 1-30.

LeFevre, K. B. (1987). Invention as a Social Act. Studies in Writing and Rhetoric by

College Composition and Communication. Carbondale, IL: Southern Illinois UP.

Lynch, M. (1982). Art and Artifacts in Laboratory Science: A Study of Shop Work and Shop

Talk in a Research Laboratory. London, England: Routledge and Kegan Paul.

Mandel, H. (1983). Funding More NIH Research Grants. Science, 221, 338-340.

Mansfield, R. S. & Busse, T. V. (1981). The Psychology of Creativity and Discovery.

Chicago, IL: Nelson-Hall.

Markle, G. E. & Peterson, J. C. (1981). Controversies in Science and Technology: A Protocol

for Comparative Research. Science, Technology, and Human Values, 6 (Spring), 25-

32.

Markus, G. (1987). Why Is There No Hermeneutics of Natural Science? Some Preliminary

Theses. Science in Context, 1, 5-51.

Marshall, C., Nelson, C., & Gardiner, M. (1987). Design Guidelines. Applying Cognitive

Psychology to User-Interface Design. M. Gardiner & B. Christie (Eds.). NY, NY:

Wiley, 221-278.

237

Martin, B. E. & Brouwer, W. (1991). The Sharing of Personal Science and Narrative

Element in Science Education. Science Education, 75 (6), 707-722.

Martin, K. & Miller, E. (1988). Storytelling and Science. Language Arts, 63 (3), 255-259.

Mattice, D. A. (1984). The Research Proposal by an Academic Institution. Journal of

Technical Writing and Communication, 4 (1), 5-17.

Mayer, R. E. (1981). The Psychology of How Novices Learn Computer Programming. ACM

Computing Surveys, 13 (1), 121-141.

Mayer, R. E. (1985). Structural Analysis of Science Prose: Can We Increase Problem-solving

Performance. Understanding Expository Text: A Theoretical and Practical

Handbook for Analyzing Expository Text. B. K. Britton & J. B. Black (Eds.).

Hillsdale, NJ: Lawrence Erlbaum, 65-88.

Mazur, A. (1981). Dynamics of Technical Controversy. Washington, DC: Communications

P.

Mehlenbacher, B. (1992). Navigating Online Information: A Characterization of

Extralinguistic Factors That Influence User Behavior. Going Online: The New

World of Multimedia Documentation. SIGDOC’92: The 10th Annual International

Conference Proceedings. NY, NY: The Association for Computing Machinery, 35-46.

Mehlenbacher, B., Duffy, T. M., & Palmer, J. E. (1989). Finding Information on a Menu:

Linking Menu Organization to the User’s Goals. Human-Computer Interaction, 4

(3), 231-251.

238

Miller, C. R. (1980). Rules, Context, and Technical Communication. Journal of Technical

Writing and Communication, 10 (2), 149-158.

Miller, C. R. (1984). Genre as Social Action. Quarterly Journal of Speech, 70, 151-167.

Miller, C. R. (1985). Invention in Technical and Scientific Discourse: A Prospective Survey.

Research in Technical Communication: A Bibliographic Sourcebook. M. G. Moran &

D. Journet (Eds.). Westport, CN: Greenwood, 117-162.

Miller, C. R. (1989). What’s Practical About Technical Writing? Technical Writing:

Theory and Practice. B. E. Fearing & W. K. Sparrow (Eds.). NY, NY: The Modern

Language Association of America, 14-24.

Miller, C. R. & Selzer, J. (1985). Special Topics of Argument in Engineering Reports.

Writing in Nonacademic Settings. L. Odell & D. Goswami (Eds.). London, England:

The Guilford P, 309-341.

Mitroff, I. I. & Chubin, D. E. (1979). Peer Review at NSF: A Dialectical Policy Analysis.

Social Studies of Science, 9 (2), 199-232.

Moran, M. G. & Journet, D. (Eds.). (1985). Research in Technical Communication: A

Bibliographic Sourcebook. Westport, CN: Greenwood P.

Mostyn, B. (1985). The Content Analysis of Qualitative Research Data: A Dynamic

Approach. The Research Interview: Uses and Approaches. M. Brenner, J. Brown, &

D. Canter (Eds.). NY, NY: Academic P, 115-145.

Mukerji, C. (1989). A Fragile Power: Scientists and the State. Princeton, NJ: Princeton UP.

239

Mulkay, M, Potter, J., & Yearley, S. (1983). Why Analysis of Scientific Discourse is

Needed. Science Observed. K. D. Knorr-Cetina & M. Mulkay (Eds.). London,

England: Sage, 171-203.

Mulkay, M. (1981). Action and Belief or Scientific Discourse? A Possible Way of Ending

Intellectual Vassalage in Social Studies of Science. Philosophy of the Social

Sciences, 11, 163-172.

Mulkay, M. & Gilbert, G. N. (1982a). Accounting for Error: How Scientists Construct Their

Social World When They Account for Correct and Incorrect Belief. Sociology, 16

(2), 165-183.

Mulkay, M. & Gilbert, G. N. (1982b). Joking Apart: Some Recommendations Concerning the

Analysis of Scientific Culture. Social Studies of Science, 12 (4), 585-613.

Mulkay, M. & Gilbert, G. N. (1982c). What is the Ultimate Question? Some Remarks in

Defence of the Analysis of Scientific Discourse. Social Studies of Science, 12 (1),

309-319.

Murphy, D. G. & Dean, D. J. (1984). Application and Review Procedures for the National

Institutes of Health Small Business Innovation Research Program. Institute

Insight: The National Institute for Entrepreneurial Technology. Washington, DC:

NIH, 1-12.

Murphy, D. G. & Dean, D. J. (1986). Biomedical Research and Research Training Support

by the National Institutes of Health. Nutrition International, 2 (1), 38-44.

240

Myers, G. (1985a). Text as Knowledge Claims: The Social Construction of Two Biology

Articles. Social Studies of Science, 15 (4), 593-630.

Myers, G. (1985b). The Social Construction of Two Biologists’ Proposals. Written

Communication, 2 (3), 219-245.

Myers, G. (1986). Writing Research and the Sociology of Scientific Knowledge: A Review

of Three New Books. College English, 48 (6), 595-610.

Myers, G. (1990). Writing Biology: Texts in the Social Construction of Scientific

Knowledge. Science and Literature Series. G. Levine (Ed.). Madison, WI: U of

Wisconsin P.

Myers, G. (1991). Stories and Styles in Two Molecular Biology Review Articles. Textual

Dynamics of the Professions: Historical and Contemporary Studies of Writing in

Professional Communities. C. Bazerman & J. Paradis (Eds.). Madison, WI: The U of

Wisconsin P, 45-75.

Nelkin, D. (1978). Threats and Promises: Negotiating the Control of Research. Daedalus,

107, 191-209.

Nelkin, D. (Ed.). (1979). Controversy: Politics of Technical Decisions. London, England:

Sage.

Nelkin, D. (1987). Selling Science: How the Press Covers Science and Technology. NY, NY:

W. H. Freeman.

Newell, A. & Simon, H. A. (1972). Human Problem Solving. Englewood Cliffs, NJ:

Prentice-Hall.

241

Niederberger, M. (1990). Science Research Funding Shrinks. Pittsburgh, PA: Pittsburgh P,

December 9.

Novello, A. C. (1985). The Peer Review Process: How to Prepare Research Grant

Applications to the NIH. Mineral Electrolyte Metabolism, 11, 281-286.

Nystrand, M. (1989). A Social-Interactive Model of Writing. Written Communication, 6

(1), 66-85.

Nystrand, M. & Himley, M. (1984). Written Text as Social Interaction. Theory into

Practice, 23, 198-207.

Odell, L. (1985). Beyond the Text: Relationships between Writing and Social Context.

Writing in Nonacademic Settings. L. Odell & D. Goswami (Eds.). London, England:

The Guilford P, 249-280.

Odell, L., Goswami, D., & Herrington, A. (1983). The Discourse-based Interview: A

Procedure for Exploring the Tacit Knowledge of Writers in Nonacademic Settings.

Research on Writing: Principles and Methods. P. Mosenthal, L. Tamor, & S.

Walmsley (Eds.). NY, NY: Longman, 221-235.

Olsen, L. A. (1989). Computer-Based Writing and Communication: Some Implications for

Technical Communication Activities. Journal of Technical Writing and

Communication, 19 (2), 97-118.

OmniProof. (1989). Caere Corporation. Los Gatos, CA: Patent 4,807,182.

Ong. W. (1975). The Writer’s Audience is Always a Fiction. PMLA, 90, 9-21.

242

Overington, M. A. (1977). The Scientific Community as Audience: Toward a Rhetorical

Analysis of Science. Philosophy and Rhetoric, 10, 143-164.

Patterson, F. & Linden, E. (1981). The Education of Koko. NY, NY: Holt.

Pearce, W. B. & Chen, V. (1989). Ethnography as Sermonic: The Rhetorics of Clifford

Geertz and James Clifford. Rhetoric in the Human Sciences. H. W. Simons (Ed.).

London, England: Sage, 119-132.

Pennington, N. (1985). Stimulus Structures and Mental Representations in Expert

Comprehension of Computer Programs. Center for Decision Research, Technical

Report No. 2-ONR, Graduate School of Business. Chicago, IL: U of Chicago.

Perkins, D. N. (1981). The Mind’s Best Work. Cambridge, MA: Harvard UP.

Petraglia, J., Flower, L., & Higgins, L. (in press). Uncovering Conceptions of Purpose at

Work in the Composition Classroom. Center for the Study of Writing Technical

Report. Berkeley, CA & Pittsburgh, PA: U of California at Berkeley and Carnegie

Mellon.

Piazza, C. L. (1987). Identifying Context Variables in Research on Writing: A Review and

Suggested Directions. Written Communication, 4 (2), 107-137.

Pinch, T. (1985). Towards an Analysis of Scientific Observation: The Externality and

Evidential Significance of Observational Reports in Physics. Social Studies of

Science, 15 (1), 3-36.

Popken, R. L. (1988). A Study of Topic Sentence Use in Scientific Writing. Journal of

Technical Writing and Communication, 18 (1), 75-86.

243

Potter, J. & Mulkay, M. (1982). Making Theory Useful: Utility Accounting in Social

Psychologists’ Discourse. Fundamenta Scientiae, 3 (4), 259-278.

Potter, J. & Mulkay, M. (1985). Scientists’ Interview Talk: Interviews as a Technique for

Revealing Participants’ Interpretative Practices. The Research Interview: Uses

and Approaches. M. Brenner, J. Brown, & D. Canter (Eds.). NY, NY: Academic P,

247-271.

Prelli, L. J. (1989a). A Rhetoric of Science: Inventing Scientific Discourse. Columbia, SC: U

of South Carolina P.

Prelli, L. J. (1989b). The Rhetorical Construction of Scientific Ethos. Rhetoric in the

Human Sciences. H. W. Simons (Ed.). London, England: Sage, 48-68.

Price, D. J. (1986). Little Science, Big Science . . . And Beyond. NY, NY: Columbia UP.

Roberts, S. (1991). Technology Transfer: An Opportunity for Technical Communication.

Technical Communication, 39 (3), 336-343.

Rorty, R. (1979). Philosophy and the Mirror of Nature. Princeton, NJ: Princeton UP.

Rose, M. (1985). Complexity, Rigor, Evolving Method, and the Puzzle of Writer’s Block:

Thoughts on Composing-Process Research. When a Writer Can’t Write: Studies in

Writer’s Block and Other Composing-Process Problems. M. Rose (Ed.). London,

England: Guilford P, 227-260.

Rothenberg, A. (1979). The Emerging Goddess: The Creative Process in Art, Science, and

Other Fields. Chicago, IL: U of Chicago P.

244

Rymer, J. (1988). Scientific Composing Processes: How Eminent Scientists Write Journal

Articles. Writing in Academic Disciplines: Advances in Writing Research, Vol. 2.

D. A. Jolliffe (Ed.). Norwood, NJ: Ablex, 211-250.

Scardamalia, M. & Bereiter, C. (1987). Knowledge Telling and Knowledge Transforming in

Written Composition. Advances in Applied Psycholinguistics, Volume 2: Reading,

Writing, and Language Learning. S. Rosenberg (Ed.). Cambridge, England:

Cambridge UP, 142-175.

Schriver, K. A. (1989a). Document Design from 1980 to 1989: Challenges That Remain.

Technical Communication, 36 (4), 316-331.

Schriver, K. A. (1989b). Evaluating Text Quality: The Continuum From Text-Focused to

Reader-Focused Methods. IEEE Transactions on Professional Communication, 32 (4),

238-255.

Schriver, K. A. (1989c). Theory Building in Rhetoric and Composition: The Role of

Empirical Scholarship. Rhetoric Review, 7 (2), 272-288.

Sebeok, T. A. (1982). The not so Sedulous Ape: Review of “The Education of Koko” by

Francine Patterson and Eugene Linden. Times Literary Supplement, 10, 976.

Selzer, J. (1983). The Composing Process of an Engineer. College Composition and

Communication, 34 (2), 178-187.

Selzer, J. (1989). Composing Processes for Technical Discourse. Technical Writing: Theory

and Practice. B. E. Fearing & W. K. Sparrow (Eds.). NY, NY: The Modern

Language Association of America, 43-50.

245

Simon, H. A. (1979). Models of Thought. New Haven, CN: Yale UP.

Simon, H. A. & Hayes, J. R. (1974). Understanding Written Problem Instructions.

Knowledge and Cognition. L. W. Gregg (Ed.). Norwood, NJ: Laurence Erlbaum, 167-

200.

Simons, H. W. (Ed.). (1989). Rhetoric in the Human Sciences. London, England: Sage.

Skolinowski, H. (1966). The Structure of Thinking in Technology. Technical Culture, 7,

371-383.

Smith, F. R. (1976). Education for Proposal Writers. Journal of Technical Writing and

Communication, 6 (2), 113-122.

Spilka, R. (1988). Studying Writer-Reader Interactions in the Workplace. The Technical

Writing Teacher, 15 (3), 208-221.

Spiro, R. J., Vispoel, W. P., Schmitz, J. G., Samarapungavan, A., & Boerger, A. E. (1987).

Knowledge Acquisition for Application: Cognitive Flexibility and Transfer in

Complex Content Domains. Executive Control Processes in Reading. B. K. Britton

and S. M. Glynn (Eds.). Hillsdale, NJ: Lawrence Erlbaum, 177-199.

Steinberg, E. R. (1986). Protocols, Retrospective Reports, and the Stream of Consciousness.

College English, 48, 697-712.

Studer, K. & Chubin, D. (1980). The Cancer Mission, Social Contexts of Biomedical

Research . London, England: Sage.

246

Suchman, L. (1987). Plans and Situated Actions: The Problem of Human-Machine

Communication. Cambridge, England: Cambridge UP.

Swales, J. (1984). Research into the Structure of Introductions to Journal Articles and its

Application to the Teaching of Academic Writing. Common Ground: Shared

Interests in ESP and Communication Studies. R. Williams, J. Swales, & J. Kirkman

(Eds.). NY, NY: Pergamon P, 77-86.

Swales, J. & Najjar, H. (1987). The Writing of Research Article Introductions. Written

Communication, 4 (2), 175-191.

Swarts, H., Flower, L. S., & Hayes, J. R. (1984). Designing Protocol Studies of the Writing

Process: An Introduction. New Directions in Composition Research. R. Beach & L.

S. Bridwell (Eds.). London, England: Guilford P, 53-71.

Thoresen, K. (1990). Experiences with Participatory Design. PDC’90: Participatory Design

Conference Proceedings. A. Namioka & D. Schuler (Eds.). Seattle, WA: 34-35.

Toulmin, S. (1972). Human Understanding: The Collective Use and Evolution of Concepts.

Princeton, NJ: Princeton UP.

Travis, G. D. (1981). Replicating Replication? Aspects of the Social Construction of

Learning in Planarian Worms. Social Studies of Science, 11 (1), 11-32.

Vande Kopple, W. J. (1982). Functional Sentence Perspective, Composition, and Reading.

College Composition and Communication, 33 (1), 50-63.

Vande Kopple, W. J. (1985). Some Exploratory Discourse on Metadiscourse. College

Composition and Communication, 36 (1), 82-93.

247

Vande Kopple, W. J. (1986). Given and New Information and Some Aspects of the

Structures, Semantics, and Pragmatics of Written Texts. Studying Writing:

Linguistic Approaches. C. R. Cooper & S. Greenbaum (Eds.). London, England:

Sage, 72-111.

Voss, J. F., Greene, T. R., Post, T. R., & Penner, B. C. (1983). Problem Solving Skill in the

Social Sciences. The Psychology of Learning and Motivation: Advances in Research

Theory, Vol 17. G. H. Bower (Ed.). NY, NY: Academic P, 165-213.

Wander, P.C. (1976). The Rhetoric of Science. Western Journal of Speech Communication,

40, 226-235.

Weaver, R. M. (1970). Concealed Rhetoric in Scientific Sociology. Language is Sermonic.

R. Johanneson, R. Eubanks, & R. Strickland (Eds.). Baton Rouge, LA: Louisiana

State UP, 139-158.

Weimer, W. B. (1977). Science as a Rhetorical Transaction: Toward a Nonjustificational

Conception of Rhetoric. Philosophy and Rhetoric, 10, 1-29.

Wells, S. (1986). Jurgen Habermas, Communicative Competence, and the Teaching of

Technical Discourse. Theory in the Classroom. C. Nelson (Ed.). Champaign, IL: U

of Illinois P, 245-269.

Winsor, D. A. (1989). An Engineer’s Writing and the Corporate Construction of Knowledge.

Written Communication, 6 (3), 270-285.

Winsor, D. A. (1990). Engineering Writing/Writing Engineering. College Composition and

Communication, 41 (1), 58-70.

248

Witte, S. P. (1987). Pretext and Composing. College Composition and Communication, 38

(4), 397-425.

Witte, S. P. & Cherry, R. D. (1986). Writing Processes and Written Products in Composition

Research. Studying Writing: Linguistic Approaches. C. R. Cooper & S. Greenbaum

(Eds.). London, England: Sage, 112-153.

Woolgar, S. (1980). Discovery: Logic and Sequence in a Scientific Text. The Social Process

of Scientific Investigation, Sociology of the Sciences Yearbook, Vol. 4. K. Knorr, R.,

Krohn, & R. Whitley (Eds.). Dordrecht: D. Reidel, 239-268.

Young, R. E. (1978). Paradigms and Problems: Needed Research in Rhetorical Invention.

Research on Composing. C. R. Cooper & L. Odell (Eds.). Urbana, IL: NCTE, 29-48.

Young, R. E. (1987). Recent Developments in Rhetorical Invention. Teaching Composition:

Twelve Bibliographical Essays. G. Tate (Ed.). Fort Worth, TX: Texas Christian

UP, 1-38.

Zappen, J. P. (1985). Writing the Introduction to a Research Paper: An Assessment of

Alternatives. The Technical Writing Teacher, 12 (2), 93-101.

Zappen, J. P. (1987). Historical Studies in the Rhetoric of Science and Technology. The

Technical Writing Teacher, 14 (3), 285-298.

Vitae