00001.tif

PDF

Download

Open document

Transcript (if available)

Content DECISION STYLE, PURPOSE OF DISCOURSE, AND DIRECTION OF COMMUNICATION: THE IMPACT ON WRITING STYLES OF SELECTED ACCOUNTANTS IN A BIG EIGHT ACCOUNTING;FIBM by Betty Parsons Pytlik A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (English) November 1982 UMI Number: D P23092 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is d e p en d en t upon the quality of the copy subm itted. In the unlikely event that the author did not sen d a com plete m anuscript and there a re m issing p ag es, th e s e will be noted. Also, if m aterial had to be rem oved, a note will indicate the deletion. UMT Dissertation Publishing UMI D P23092 Published by P roQ uest LLC (2014). Copyright in the D issertation held by the Author. Microform Edition © P roQ uest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United S ta te s C ode ProQuest P roQ uest LLC. 789 E ast Eisenhow er Parkw ay P.O. Box 1346 Ann Arbor, Ml 4 8 1 0 6 - 1346 UNIVERSITY O F S O U T H E R N CA LIFO RN IA T H E G R A D U A T E S C H O O L U N IV E R SIT Y PA R K L O S A N G E L E S . C A L IF O R N IA 9 0 0 0 7 This dissertation, written by BETTY PARSONS PYTLIK under the direction of h.ex.... Dissertation C o m mittee, and approved by all its members, has been presented to and accepted by The Graduate School, in partial fulfillment of requirements of the degree of D O C T O R O F P H I L O S O P H Y Dean D ate January S, 1983 DISSERTATION COMMITTEE Chairman ii To Doug ACKNOWLEDGMENTS In retrospect, it seems as if a cast of thousands assisted and encouraged me while I developed the methodology for this study, collected and analyzed the writing samples, and wrote and re-wrote about what I'd found. During that process, my committee members, Professors W. Ross Winterowd, Betty Bamberg, and J. Douglas Andrews, were always accessible and helpfully critical. Professor Robert Kaplan, in the Department of Linguistics, also was accessible to answer questions and, more frequently, to ask "So?" and "Why not?" My colleagues and friends at the American Language Institute, in the Freshman Writing Program, in the Rhetoric, Linguistics, and Literature Program, and in the Business Communication Department were incalculably helpful to me. They listened, critiqued, rated memoranda, proofread, encouraged me, and didn't give up on me when I failed to return their calls. Without the contributions of the accounting firm, of course, I could not have done the study. I thank the partner for his confidence in my project, his secretary for collecting the preliminary materials and memoranda, and the accountants for their time and their memoranda. Finally, I hope that Bob Bursick, Liz Lynn, Doug Andrews, and Paul Ballard know how important their friendships were to me while I worked on this dissertation and are to me now. iv. TABLE OF CONTENTS , Page ACKNOWLEDGMENTS...................................................iii LIST OF FI G U R E S....................................................vi LIST OF TABLES ....................................................vii CHAPTER I. STATEMENT OF PROBLEM ..................................... 1 Background and Justification .................... 1 The Purpose of the Study................................26 Limitations and Delimitations of Study ............. 27 The Plan of the Dissertation............................28 II. REVIEW OF THE LITERATURE...................................29 Nonacademic Written Discourse ...................... 29 Purpose in Written Discourse ......................... 34 Reader-Writer Relationship in Written Discourse . . 37 Decision Style........................ 39 III. METHODOLOGY................................................ 43 Collection of Writing Samples ....................... 43 Identification of Memoranda Purposes ................ 46 Coding Stylistic Features ........................... 50 Statistical Analysis .................................. 52 IV. DECISION STYLE AND WRITING STYLE . ..................... 53 Cohesive Elements .... ........................... 55 Elements of Coherence.................... 58 Features That Imply Reader-Writer Relationships . . 60 Conclusions..............................................63 V. PURPOSE AND WRITING S T Y L E .................................65 Cohesive Elements .................................... 65 Elements of Coherence.......................... .. . 68 Features That Imply Reader-Writer Relationships . . 70 Conclusions..............................................75 Page VI. DIRECTION OF COMMUNICATION AND WRITING STYLE .......... 78 Cohesive Elements .................................... 79 Elements of Coherence ................................ 87 Features That Imply Reader-Writer Relationships . . 91 Conclusions .......................................95 VII. SUMMARY OF FINDINGS, CONCLUSIONS, AND IMPLICATIONS FOR TEACHING AND RESEARCH................................97 The Context for Written Discourse .................. 98 Linguistic Choices........... 100 Purpose and Writing Style ........................... 120 Direction of Communication and Writing Style .... 125 Decision Style and Writing Style .................. 126 Summary of Study........................................ 127 APPENDIXES........... 130 A. Proposal to Accounting F i r m ............................. 130 B. First Letter to Accountants ............................ 132 C. Rowe's Decision Style Inventory ....................... 133 D. Fact Sheet for Accountants . . .........................135 E. Second Letter to Accountants ............................ 136 F. Facts About the Participating Accountants ............ 138 G. Illustrations of the Five Categories of Memoranda Purposes................................................. 141 H. Instructions to the Memoranda Readers ................ 146 I. Example of Coded Memorandum ............................ 147 N O T E S .................................... 149 BIBLIOGRAPHY ........................................................ 161 ■-•vi LIST OF FIGURES Figure Page 1. Summary Decision Style Chart ........... . 15 2. Cognitive-Contingency Decision Style Model: Environmental Complexity ............................... 18 vi i LIST OF TABLES Table Page 1. Number of Memoranda According to Dominant and Back-Up Decision Styles ........................... 45 2. Memoranda Classified According to Direction of Communication as Identified by Writers ............. 47 3. Purposes of Memoranda Identified by Three Trained Readers.................................................... 50 4. Stylistic Features Coded ..................... 51 5. Decision Style and Cohesive Elements: Frequencies of Cohesive Elements in Memoranda in T-Units .... 55 6. Decision Style and Cohesive Elements: Frequencies of Elements of Coherence in Memoranda in T-Units . . 58 7. Decision Style and Direction of Communication: Frequencies of Reader-Writer Relationship Items in Memoranda in T-Units . . .................... 61 8. Purpose and Cohesive Elements: Frequencies of Cohesive Elements in Memoranda in T-Units ........... 66 9. Purpose and Cohesive Elements: Frequencies of Elements of Coherence in Memoranda in T-Units .... 69 10. Purpose and Direction of Communication: Frequencies of Reader-Writer Relationship Items in Memoranda in T - U n i t s........................................ 71 11. Direction of Communication and Cohesive Elements: Frequencies of Cohesive Elements in Memoranda in T-Units ................................................80 12. Direction of Communication and Elements of Coherence: Frequencies of Elements of Coherence in Memoranda in T - U n i t s ................................................90 13. Direction of Communication and Reader-Writer Relationships: Frequencies of Reader-Writer Relationship Items in Memoranda in T-Units ......... 92 viii Table Page 14. Cohesive Elements: Reference Items Per T-Units .... 102 15. Cohesive Elements: Grammatical Subjects Per T-Units . 104 16. Cohesive Elements: Kinds of Verbs and Voices Per T-Units................................................ 105 17. Elements of Coherence: Conjunctions Per T-Units . . . 108 18. Elements of Coherence: Sentence Openers Per T-Units . 109 19. Reader-Writer Relationship Items: Hedges Per T-Units................................................ 113 20. Reader-Writer Relationship Items: Modals Per T-Units........................•. < ............. 114 21. Reader-Writer Relationship Items: Personal Pronouns Per T-Units ................. '.............. . 115 22. Reader-Writer Relationship Items: Formulary Expressions Per T-Units........................ 116 23. Discourse Length and Decision Style .................. .118 24. Discourse Length and Purpose of Discourse . ......... 119 25. Discourse Length and Direction of Communication . . . . 119 26. Purposes of Memoranda Identified by Three Trained Readers and the Writers of the Memoranda..........121 1 CHAPTER I STATEMENT OF THE PROBLEM Background and Justification Communication researchers and teachers have regularly questioned leaders in business and industry about writing skills required on the job. In a recent study by Donna Stine and Donald Skarzenski, for instance, business executives and faculty were asked to identify what skills college writing courses should teach future employees, what kinds of writing employees do most-often, what problems appear in writing that crosses their desks, and how important writing is for job advancement. In another survey conducted by Paul V. Anderson, over a thousand alumni of Miami University were asked what kinds of writing they do, how important writing is, and how much writing they do on the job. And in a 1982 survey, Lester Faigley and colleagues at the University of Texas in Austin asked college-trained respondents to assess the importance of writing abilities in the world of work and to identify the types of writing done on and off the job, the media they use, and future writing needs of college-trained people.1 Surveys have also been conducted to assess communication needs in specific professions. J. Douglas Andrews and Robert J. Koester, for example, asked accounting educators and recently graduated 2 accountants what communication skills accounting professionals expect of newly hired employees.2 While surveys have identified communication needs in the world of work, intention, audience, rhetorical strategies, and style in nonliterary and nonacademic discourse have seldom been studied. In fact, only within the last ten years have discourse theorists and practitioners acknowledged that working adults produce a great deal of written discourse. Only recently, then, have they observed that we know little about what tasks working adults perform in their on-the-job writing, how the writers construe their readers, how their perceptions of audience affect their style, what stylistic features would help specific audiences in business to process information most efficiently, and what stylistic choices working writers make.3 We need to know what makes certain language options effective or ineffective for a given purpose and context so that we can begin to measure effectiveness. First, however, we must be able to describe discourse generated in specific contexts, for specific purposes, and for specific readers. To this end, this study describes style in discourse written by accountants for other accountants within one firm. Writing Style Any quantitative study of style, whether in literary, academic, or nonacademic discourse, should aspire to the methodological rigor demonstrated in Louis T. Milic's A Quantitative Approach to the 3 Style of Jonathan Swift.5 Underlying Milic's study are assumptions about a writer's style that also underlie this study: . . . the writer must respond to the necessity of presen ting material in a persuasive or moving way; or he must find a logical sequence to impose on the disorderly array of facts he has collected. His response to these chal lenges is always conditioned by his culture, the choices available in his language in his time and place. But it is also limited by the choices he makes deliberately for the effect on the reader. And finally it is given form by those he is impelled to make by some unconscious predisposition.6 Furthermore, Milic suggests, a style "contains a uniform and constant diffusion of [a writer's] mind and personality, expressed through certain grammatical categories which may be measured objectively." In a writer's grammar, claims Milic, "lies the key to his style, provided the proper categories for investigation can be developed."7 The same degree of rigor is achieved in Mary Hiatt's The Way Women Write, in which she examines sentence length, sentence complexity, parallel constructions, rhetorical devices, similes, adjectives, and adverbs in fiction and non-fiction. Hiatt characterizes as narrow Milic's view that style is totally a matter of one individual's writing, given the evidence of shared stylistic and linguistic characteristics of, for instance, the "Attic" or "Baroque" writers. She concludes that one can claim that an individual's style can be deemed individual only as it varies from a group style. So, she set out to see if a group style exists, asking "is there a feminine style?" She found that the profile of women's 4 writing is indeed different from that of the men's. The style of the women writers is rhetorically more effective at the same time that it is more conservative in other aspects of expression. It is, in general, a middle-of-the-road style, not given to extremes of length or brevity, not given to extremes of emotion or action, not given to extremes of "feminine" concerns to the exclusion of "masculine" concerns.8 This analysis of style in the memoranda of accountants is influenced generally by the work of M. A. K. Halliday and Ruqaiya Hasan. In "The Linguistic Study of Literary Texts," M. A. K. Halliday claims that linguistics can describe all written texts.9 Referring to this textual analysis as "linguistic stylistics," Halliday observes that "A text is meaningful not only in virtue of what it is but also in virtue of what it might have been."10 Linguistic stylistics, then, is a comparative study. Halliday illustrates his point by contrasting the distribution of verbal items and the structure of nominal groups in two poems to show that a narrative and a sonnet, a defining form, differ in linguistic properties. The value of linguistic stylistics to the present study lies in its ability to compare features. All of the stylistic features analyzed in this study are discussed to some extent in Halliday's important three-part Journal of Linguistics article, "Notes on Transitivity and Theme in English."11 In the article, Halliday characterizes transitivity in terms of the type of process ("doing" and "being" are the two general processes) expressed in a main clause, the participants in 5 the process, and attributes and circumstances of the process and participants. Halliday, like Charles Fillmore,12 is concerned with the semantic relationships among actions, actors, goals, and attributes and, again like Fillmore, classifies verbs according to the clause features which can occur with the verbs. Halliday concludes that some role relationships take precedence over others when they are mapped in the surface grammar. For example, in English, unless special conditions of staging and informational coherence13 come into play, the agent of a proposition matches the surface subject. If there is no agent, the patient matches the subject. In other words, the speaker or writer chooses one element that he is referring to as the point of departure for the relationship to all other elements. This choice determines word order.14 Cohesive Features of Style In Cohesion in English, which represents the first major attempt to explain and code the cohesive elements in English, Halliday and Ruqaiya Hasan say that cohesion occurs where the INTERPRETATION of some element in the discourse is dependent on that of another. The one PRESUPPOSES the other, in the sense that it cannot be effectively decoded except by recourse to it. When this happens, a relation of cohesion is set up, and the two elements, the presupposing and the presupposed, are thereby at least potentially integrated into a text.15 The concept of cohesion, then, is semantic; it "refers to relations 6 of meaning that exist within the text, and that define it as a text.”16 Reference. One category of cohesive devices that Halliday and Hasan discuss is reference: "instead of being interpreted semantically in their own right, [reference items] make reference to something else for their interpretation."17 In this study, these categories of reference are examined: Pronouns, possessive determiners, demonstratives, and the definite article the. Nominalization. In "Nominal and Verbal Styles," Rulon Wells defines a nominal style as "the tendency to use nouns in preference to verbs."18 Wells notes that most appraisers of style deem the nominal style inferior to the verbal style. Those who judge the nominal style inferior say that nouns are more static, less vivid than verbs; that a nominal style creates longer sentences, which are generally less vivid and less comprehensible than shorter ones; and that a nominal style limits syntactic options, thus creating a monotonous style. Negative reactions to highly nominalized prose are not based solely on impressionistic responses. There is, of course, psycholinguistic evidence that a nominal style, which Rosemary Hake and Joseph M. Williams define as a style that relies heavily on nouns that have underlying expressible verbs or adjectives, can make prose difficult to process.13 Nevertheless, writers choose to use nominalized prose for several reasons: to hide an absence of 7 content, to emphasize an idea, or to help the reader move smoothly from one idea to another (that is, to achieve cohesion).20 In addition, Wells points out, some writers believe that a nominal style is easier to write, that it is helpful in achieving impersonality, avoiding commitment, and setting off the writing as esoteric, specialized, and technical.21 Nominalizations, or nominal forms, are also cohesive devices, in that writers frequently use them to connect grammatical subjects to their preceding referent. The definition of nominalization is adopted from one that Rosemary Hake and Joseph M. Williams propose in "Style and Its Consequences: Do . As I Do, Not As I Say": A nominalization is a noun that has an underlying expressible English verb or adjective. Forms such as stone and whip that have related verbs stone and whip are rejected as derived nominaliza- tions on the grounds that they do not occur in construc tions of this sort: his stone/whip of the man as opposed to : his discovery /punishment of the man.22 Example: An additional adjustment was made to the January 31, 1982 balances that should be posted to the December 31, 1981 trial balance. Gerunds are also included in this category of nouns. Grammatical Subjects. The grammatical subject of a clause is the noun that takes the subject position in a Subject-Verb-Object language like English. Wallace Chafe, in "Givenness, Contrastive ness, Definiteness, Subjects and Topics," distinguishes grammatical subjects from logical and psychological subjects: 8 A typical way of demonstrating the independence of these three roles is to point out that in The onions were peeled by Betty, although Betty remains the logical subject, the other two roles are taken by the onions. In The onions, Betty peeled, furthermore, it may be said that the onions is the psychological subject, whereas Betty is the logical and grammatical one.23 Passive Voice Verbs. There are two kinds of passives in English. In one, as Grimes puts it in The Thread of Discourse, "the agent who instigates an action is expressed in a prepositional phrase and some other element of the action is the grammatical subject."24* In the other type, the nonagentive, the agent of the action is omitted. Because the passive voice reverses the usual order of agent- action-goal, it requires more time for the reader to process than an active voice structure does.25 Still, writers choose to use the passive voice for a variety of reasons: It may simply be a means to achieve sentence variety. It also allows the writer to sidestep the question of who instigated something, and it deletes the agent when the agent is less important than the action or goal. Additionally, the passive can also be used to achieve cohesion and to help make information salient.26 By analyzing the writer's verb choices, we can determine the kinds of information the writer chooses to emphasize, what he assumes about his reader's knowledge of the subject of the discourse, and whom he identifies as the agent (if he identifies an agent). 9 Agents as Subjects. Comprehensibility of prose can be increased if agents are the grammatical subjects of sentences.27 Prose in which agent and subject are matched is easier to process because the reader knows where the writer is leading him; he can make sense of what is to come and integrate the details in the text with the writer's intention. In Revising Functional Documents: The Scenario Principle,28 Flower, Hayes, and Swarts report that in comparing original and revised public documents, they found that prose which told the writer his or her part in the action was easier for the subjects to understand. Sentences that had human agents as subjects were even more easily understood. Elements of Coherence Another feature of discourse is coherence, which, as H. G. Widdowson explains in Teaching Language as Communication, has "to do with the illocutionary function of propositions, with how they are used to create different kinds of discourse."29 A discourse, he says, is coherent "to the extent that we recognize it as representative of normal language use, to the extent that we can accept the sequence of illocutionary acts as conforming to known conventions."3 0 Conjunctions. In this study, logical relationships between propositions are identified by the conjunctions that connect them; specifically, they are identified by the four types of conjunctions that Halliday and Hasan describe in Cohesion in English. 10 Conjunctions, they say, are words or phrases that specify "the way in which what is to follow is systematically connected to what has gone before."3 1 Additives are conjunctions that add something to what has been stated. Example: Gary and I have reviewed copies of all recent reports with errors, and it appears that the major error appears to be incorrect contract numbers. Adversatives are words or phrases that express the adverse of the previous sentence or sentences. Example: I feel the letter should be issued to X, as the home office; however, I am not certain to whom specifically. Causals signal result, reason, and purpose relationships. Example: Due to an extremely tight year-end timetable, the time and expense reports for the eleven (11) days ending June 30, 1981 must be submitted for review by 10:00 a.m., Monday, June 29. Therefore, it is necessary to estimate your hours for Monday (June 29) and Tuesday (June 30) . Temporals are words or phrases that connect successive events or phenomena. Example: Last year, to alleviate the cash drain of unprofitable operations, X markedly reduced its work force and began closing eastern region plants. 11 Sentence Openers. Because of their initial positions in sentences, sentence openers contribute to coherence. Classified by Francis Christensen into three groups, they are words, phrases, and clauses that "stand before the subject, but do not form a grammati cal unit with it, whether the subject is a single word . . . or the head of a noun phrase . . . ,"32 Christensen lists four kinds of adverbial sentence openers: 1. Clauses: Example: Should they select the manual system, I can continue as planned with the set-up of the training as my audit schedule permits. 2. Prepositional Phrases: Example: For your convenience, a copy of the 1980 Summer schedule is also attached. 3. Adverbs: Example: Accordingly, copies of all P-66's that are due and that have not yet been submitted to the personnel department are to be sent to the evaluated staff person's management counselor no later than Friday February 27, 1981. (The latter underlining appears in the original.) 4. Nouns in an Adverbial Function: Example: Each year, we have made adjustment for X's cutroff of tank car purchases. 12 Christensen lists five kinds of verbal sentence openers: 1. Present Participles: Example: Regarding the $1,200,000 adjustment to inventory at January 31, 1982 that I referred to in my memo, the client has not provided us with the data needed to rollback this adjustment. 2. Past Participles: Example: Based on the results of the work done at X , we are satisfied that the working papers support the inclusion of the X trial balance in the X financial statements. 3. Infinitives: Example: To meet this goal, X has arranged for a Japanese company to manufacture beds which X would then purchase and sell to Y possibly commencing in October. 4. Absolutes: Example: The really critical problems solved, they . . . . 5. Gerunds: Christensen uses "gerunds" to classify verb + ing forms preceded by a preposition, such as After seeing his mother, he . . . . In this study, sentence openers classified as gerunds are simply verb + ing forms that appear at the beginning of sentences. Example: Proper reporting of time and expense is extreme ly important. Five inverted constructions are identified by Christensen: 1. Inverted Appositive Nouns (Noun Clusters): 13 Example: The mother of three children, she .... 2. Inverted Appositive Adjectives (Adjective Clusters): Example: Angry at the long series of delays, the men .... 3 . Inverted Progressive Forms: Example: Following is a summary of items requiring consideration as noted during review of X workpapers for their September 30, 1981 audit of Y, Inc. 4. Inverted Passive Voice Form: Example: Attached is a copy of the blank schedule for July, August and September 1981. 5. Inverted Complements: Example: What he wanted he usually got. Decision Style Just as one's writing style is conditioned by the constraints Milic identifies--by context, choices available in the language, the effect intended, and by some unconscious predisposition--so too is one's decision making style. Decision style is "the manner in which managers think and react to problems, the way they perceive, their cognitive response, their values and beliefs, and their handling of pressures."33 Through analyses of decision styles, we can specify the amount and kind of 14 information most likely to be preferred by managers and better understand how managers solve problems, react to subordinates, and respond to motivational factors and to stress.34 After a manager’s decision style preferences have been identified, they can be matched with the demands of an organization, peer pressure, task requirements, and his self definition. It is this match that produces effective decision making: it allows managers to deal more effectively with organizational problems.35 Decision styles can be identified by a variety of inventories based on several models of decision making processes. One description of the decision making process, noteworthy here because of its implications for business communication, is the model developed by Michael Driver.36 The attributes of the four dominant decision' styles that Driver identifies--Decisive, Flexible^ Hierarchic, and Integrative--are summarized in the chart below. As the last category on the chart shows, preferences for particular features of communications vary with the decision styles. 15 FIGURE 1. SUMMARY DECISION STYLE CHART” Decisive Flexible Hierarchic Integrative Values Efficiency Speed -Consistency Adaptability Speed Variety Quality Rigorous Method System Information Creativity Planning Low data base Short-range Tight control for results Low data base Intuitive High data base Long-range Tight control of method and results High data base Long-range Adaptive Goals Few; organiza tion focus Many; self- focus Few; self-focus Many; self- and organization- focus Organization Short span of control Rules Classic organizat ion Control by confusion Loose Wide span of control Elaborate procedures Automation Team process Matrix organization Communication Short summary One solution Short summary Multiple solutions Long, elaborate reports Problem, methods, data, give "Best Conclusion" Long, elaborate Problem analysis from many views; Multiple solution Among the most frequently administered decision style inventories is the Myers-Briggs Type Indicator (MBTI), which measures cognitive styles using a Jungian typology. As Robert W. Zmud explains in "On the Validity of the Analytic-Heuristic Instrument Utilized in 'The Minnesota Experiments,"' [It] provides continuous and largely independent measures on whether (1) perception is accomplished through sensing or intuition (SN), (2) judging is accomplished through thinking or feeling (TF), (3) the 'field of view' is that of an extravert or an introvert (El), or (4) judging or perception (JP) is preferred as a 'way of life'. 38 16 Driver has also developed and validated an administrative problem solving exercise, Integrative Style Test (1ST). 1ST is a computer-scored test which describes a fictitious business problem and presents six informational statements of positive and negative qualities and a series of questions that require the test-taker to take a stand on the problem. To measure cognitive flexibility, the subject is then given two new pieces of information and asked to reconsider the original answers.39 Driver has inferred preferences for particular features in organizational communications from the data collected in his validation of the 1ST. However, Alan J. Rowe's Decision Style Inventory III40 was used in this study because, unlike other decision style measures, it is based on cognitive complexity and environment, sharing with composition research a concern for both cognitive styles of writers and the rhetorical situations in which writers compose. Like decision making situations in a business environment, a rhetorical situation is constituted by exigence, audience, and constraints. An exigence, Lloyd Bitzer says, "is an imperfection marked by urgency," and it can be modified only through discourse. The second constituent is the audience, "those persons who are capable of being influenced by discourse and of being mediators of change." Also contained in a rhetorical situation "is a set of constraints made up of persons, events, objects, and relations which 17 are parts of the situation because they have the power to constrain the decisions and actions needed to modify the exigence."1 *1 t Responses on Rowe's 20-item self-scoring questionnaire reflect the subject's preferences or reactions to given situations: the score "reflects how you see yourself, not what you believe is correct or desirable, as related to your work situation."42 Like Driver's 1ST, Rowe's inventory identifies four categories of decision-making styles: analytic, conceptual, directive, and behavioral. It is based on the "cognitive-contingency" model (see chart below), which considers the manager's cognitive complexity (his ability to perceive and understand information) and environmental complexity (his concern for technical vs. sociotech- nical systems). Most decision style theorists agree that people use two or more styles when they make decisions. Thus, Rowe explains, "the manager concerned with strategic decisions tends to use the analytic and conceptual styles, whereas the operating manager concerned with performance on the job is more prone to use the directive and behavior styles."43 In this study, decision styles were identified as dominant and back-up, e.g., Analytic-Conceptual. 18 FIGURE 2. COGNITIVE-CONTINGENCY DECISION STYLE MODEL'*'1 ENVIRONMENTAL COMPLEXITY Cognitive Complexity Technical Sociotechnical Ambiguity Analytic Conceptual Structured Directive Behavioral In Strategic Management and Planning: A Methodological Approach, Rowe and his co-authors describe the four decision styles.**5 In the summaries below, generalizations about managers’ dominant decision styles have been abstracted on the basis of their relevance to this study of writing style. Directive Style: Has low tolerance for ambiguity, needs structure; Uses little information and few alternatives; Has need for power, status, security; Is systematic, but aggressive; Prefers verbal communication. Analytic Style: Has greater tolerance for ambiguity than Directive does; Is problem-solving oriented; Values results; 19 Prefers written communication. Behavioral Style: Is receptive to suggestions, but needs structure; Is supportive of subordinates; Is persuasive; Seeks acceptance and avoids conflict; Prefers verbal communication. Conceptual Style: Has high tolerance for ambiguity; Uses much data and many alternatives; Is perfectionist, values quality; Prefers written reports. Rowe's Decision Style Inventory, then, specifically designed to elicit decision style preferences in business environments, was used in this study for both theoretical and practical reasons: In addition to being based on cognitive complexity and environment, thus sharing with composition research a concern for both cognitive styles of writers and the rhetorical situations in which writers compose, Rowe's measurement has been shown to be more discriminating than other instruments .**6 Furthermore, the Rowe measurement can be completed quickly. Given the time constraints under which the accounting subjects operate, Rowe's instrument was the most practical to administer. Purpose in Discourse One assumption that underlies rhetorical theory and modern communication theory is that, as Lee Odell has expressed it, 20 "Writers must have a good sense of the purpose they wish to accomplish in their writing."'*7 Another assumption is that the writer's purpose must be made clear to the reader.1 *8 In the literature, a variety of terms is used interchangeably to discuss writers' purposes. E. D. Hirsch, for example, considers the writer's "intention": "The quality of the text," Hirsch says, "is judged according to its success in fulfilling its own implicit intention.u9 James Britton et al. identify expressive, transactive, and poetic discourse as "function categories."50 Frank D'Angelo retains the term "rhetorical categories," or modes.51 However, the most frequently used term, and the one used in this study, is "purpose." For James Kinneavy, purpose in discourse determines everything else in the languaging process.52 Accordingly, Kinneavy classifies discourse as expressive, literary, persuasive, and referential. Lee Odell reminds us that "Writing tasks involve appeals to different audiences and accomplish different purposes."53 De Beaugrande cautions that "Before we can prescribe or proscribe certain language options, we must discover what makes them effective or ineffective for a given purpose and context."51* In this study, the purpose of each writing sample was identified in terms of illocutionary intention and illocutionary effect. An illocutionary act, defined in John L. Austin's How To Do Things With Words and elaborated upon in John Searle's Speech Acts: An Essay in the Philosophy of Language, "A Classification of 21 Illocutionary Acts," and "What Is a Speech Act?" is an act of doing, such as stating, requesting, confirming, promising. Illocutionary intention is the expectation the writer holds when, for example, he proposes an action to the reader: The writer expects the reader to perceive his or her intent. An illocutionary effect occurs when the reader recognizes the writer's intent.55 Reader-Writer Relationship Until recently, most sociolinguists, who examine the speech act in all its social dimensions,56 have focused on speakers and hearers in conversational sequences. John Searle, William Labov, C. A. Ferguson, Dell Hymes, H. P. Grice, J. M. Sinclair and Malcolm Coulthard, and M. A. K. Halliday, Angus McIntosh, and Peter Strevens,57 for example, consider how speaker-hearer-context relationships are manifested in conversations. These and many other linguists and language philosophers have shown that one can infer speakers' attitudes toward their listeners from their use of such linguistic structures as questions, politeness formulas, and hedges and choices of modals and pronouns.5 8 More recently, conversational analysis techniques have been applied to written discourse. Mary Louise Pratt, Stanley Fish, and Olga Garnica, to name only a few, have applied speech act theory to literary analysis.59 And stylistic features of nonliterary and nonacademic discourse have also been examined. David Crystal and Derek Davy, for instance, have examined newspaper and legal writing;60 David Mellinkoff has examined legal writing;61 and as the 22 extensive bibliography in Document Design; A Review of the Relevant Literature6 2 indicates, the Document Design Center has used discourse analysis techniques to analyze public documents so they can be rewritten into more easily comprehensible prose. Especially important to this study of reader-writer relationships are the observations of George Lakoff, Robin Lakoff, and Roger Brown and Albert Gilman, although it should be noted that their respective studies examine spoken discourse, not written. In "Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts," George Lakoff discusses degrees of truth in natural language. Natural language concepts, he says, "have vague boundaries and fuzzy edges and . . . consequently, natural language sentences will very often be neither true, nor false, nor nonsensical, but rather true to a certain extent and false to a certain extent, true in certain respects and false in other respects."63 This vagueness and fuzziness, Lakoff says, can be studied methodically, and he posits four kinds of "hedges" that are context-free, words and phrases whose truth values are dependent on, for example, definition. There are, additionally, hedges that depend on context. 6,1 The semantic operation of hedging can be achieved in a variety of forms. Moreover, the presence or absence of hedges and the preference of writers for specific hedges are matters of style and considered in the proposed study. 23 In "The Logic of Politeness: Or Minding Your P's and Q's," Robin Lakoff looks at situations in which communicating an important idea is secondary to reaffirming or strengthening relationships: The rules of politeness--Don't impose, Give the addressee options, Make the addressee feel good--supercede the conversation maxims formulated by H. P. Grice.65 Some incidences of politeness are hedges, so they are examined in this study since they reflect the speaker's attitude toward his social context: more specifically, his assumptions about (1) the people he is communicating with: their feelings about him, their rank relative to his; (2) the real- world situation in which he is communicating: how crucial is the information he seeks to convey? does he seek to convey information? how formal is the situation of the speech act? and (3) his decisions based on (1) and (2) as to the effect he wished to achieve via his communicative act: does he want to reinforce the status distinction between himself (SP) and the addressee (A)? To obliterate them?66 Two classes of hedges, Approximators and Shields, have been identified by Ellen Prince, Joel Frader, and Charles Bosk.67 Approximators, of which there are two types, introduce fuzziness within the propositional content proper. 1. Adaptors. In situations in which writers use Adaptors, they assume that their reader and they agree on the characteristics of the referent and that they share a lexicon to describe that referent. Example: . . . and everyone has reasonably free access to program documentation. 24 2. Rounders. Rounders are used to convey a range of a term, "the term given being typical or exemplary of the range and the hedge indicating that it is in fact a range."68 Example: Although we resolved his questions on the system some 6 or 8 weeks ago, Joe has not yet placed the order for the forms so that we could begin training/set up. Shields, of which there are two types, reflect writers' person al commitment to the truth of the propositions they are conveying. 1. Plausibility Shields. A Plausibility Shield indicates that the writer is less than fully committed to the truth of the proposition conveyed. Example': I understand that they are needed for a planning meeting with the client. 2. Attribution Shields. An Attribution Shield "attributes the belief in question to someone other than the speaker, the speaker's own degree of commitment being only indirectly inferable.69 Example: It is estimated that the observation (including inventory, observation questionnaire write-up, test counts, receiving and shipping cutoff, etc.) will require approximately 8 hours of a staff accountant's time. In another important article, "The Pragmatics of Modality," Lakoff concludes that a speaker's choice of modals is based on real- world situations and social and other contextual assumptions. In her consideration of can and may and must and should, Lakoff is concerned with the contextual 25 assumptions that are shared by speaker and addressee, whether or not previously given linguistic expression in the discourse; the social situation assumed by partici pants in the discourse; the impression the speaker wants to make on the addressee.70 The present study is not concerned with the "correctness" and "appropriateness" of modals in the selected memoranda; rather it is concerned with which modals appear in the memoranda written in "real-world situations" with "social and other contextual assumptions that constrained the writers and guided their choices of modals." In Modality and the English Modals, F. R. Palmer identifies the following modals, all of which suggest possibility or necessity:71 used to will dare shall need can is to may must ought to In their 1960 essay, "The Pronouns of Power and Solidarity,"72 Roger Brown and Albert Gilman trace the history of the use of jtu and vous, concluding that the social relationships of power and solidarity determine the choice of pronouns. The relevance of Brown and Gilman's study to the present study lies in its suggestion that choices of pronouns can be indicators of power and solidarity. 26 The Purpose of the Study The purpose of this study is to determine what relationships exist between the decision-making strategies and the writing strategies of selected accountants. The problem is (1) to discover what stylistic features appear in a stratified sample of intraorga- nizational writing from selected accountants and (2) to compare features in the writing of accountants identified as preferring the analytic style in decision making with features in the writing of accountants preferring conceptual, directive, or behavioral styles. These questions were posed: 1. Do writers' decision style preferences affect their choices of cohesive elements, sentence openers, transitions, hedges, modalities, and personal pronouns? 2. Does the purpose of the communication, that is, the task, affect a writer's choice of cohesive elements, sentence openers, transitions, hedges, modalities, and personal pronouns? 3. Does the direction of the communication affect a writer's choice of cohesive elements, sentence openers, transitions, hedges, modalities, and personal pronouns? 27 Limitations and Delimitations of Study I attempted to obtain a writing sample representative of a cross section of decision stylists from within one accounting office; this was difficult to achieve for two reasons. First, some dominant/back-up decision styles are not represented among the middle-level accountants in the participating firm. Perhaps the nature of accounting tasks does not attract certain decision stylists and perhaps some decision stylists not represented in the sample have assumed other positions with the firm. For example, they may have moved into Management Advisory Service (MAS), into the tax department, or into senior management levels. Second, as some participating accountants noted on their fact sheets, the nature of accounting tasks frequently makes oral communication, rather than written communication, more efficient and expedient. Nevertheless, because the samples represent a variety of writing purposes and are addressed to a variety of readers, observations about the stylistic features will be generalizable to other written discourse in accounting and to functional/transac tional written discourse in other business contexts. In addition to the limit that was placed on the study because of the nature of the accounting profession, I also imposed several limits. First, because studies of writing in accounting generally emphasize the need for improved communication among and from 28 mid-level accountants, I asked the partner of the firm to solicit memoranda from mid-level accountants only. I also asked that the readership of the memoranda be exclusively within the firm; thus, the writers and readers would share a well-defined rhetorical context. Finally, the linguistic features analyzed are the sentence and intersentence levels. No attempt was made to analyze the macrostructure of the discourse. The Plan of the Dissertation In Chapter II, literature relevant to nonacademic discourse, purpose, reader-writer relationships, linguistic analysis of style, and decision style is reviewed. The methodology of the study is described in Chapter III. Summaries of the findings of the study and analyses of stylistic preferences of decision stylists and stylistic features associated with the purposes of the discourse and with the relationship between the reader and writer are in Chapters IV, V, and VI. Implications of the study are discussed in Chapter VII: Pedagogical suggestions for developing rhetorical skills of accountants and on-the-job writers in general are made, and direc tions for future research in nonacademic discourse are identified. 29 CHAPTER II REVIEW OF THE LITERATURE In this review, I summarize studies about nonacademic written discourse, studies that have contributed to our understanding of how purpose and reader affect style, and literature that relates deci sion making styles to writing styles. Nonacademic Written Discourse The need to produce public and institutional documents comprehensible to the ordinary reader is recognized in such diverse places as the composition classroom, the White House, state legislatures, the courts, and the boardroom.73 Demands for readable prose in government forms, guarantees, leases, insurance policies, and advertisements are being met by interdisciplinary efforts like those of the Document Design Project. Begun by the American Institutes for Research in 1978, the Project is charged with the task of increasing the knowledge and skills of people who produce public documents. As Daniel B. Felker explains in Document Design: A Review of the Relevant Literature, the Project conducts "theoretical and applied research studies on language comprehension, on the ways in which skilled and unskilled writers work, [and] on problems associated with different document features"; it works "directly with government and private agencies as they produce 30 materials for public use"; and it conducts "courses on writing and design for graduate students and undergraduates."7* Reports of studies conducted by the Document Design Center comprise a large portion of the current literature on interdisci plinary attempts to assess and meet the needs of the reader. Representative of these reports--and especially pertinent to the present proposed study--are Linda Flower, John Hayes, and Heidi Swarts's Revising Functional Documents: The Scenario Principle, Marshall A. Atlas's Addressing An Audience: A Study of Expert- Novice Differences in Writing, Andrew M. Rose and Louis A. Cox Jr.'s Following Instructions, and Daniel B. Felker et al.'s Guidelines for Document Designers.75 The goal of Flower, Hayes, and Swarts's study was to "define a set of powerful principles which would allow writers to revise public documents so that they meet the needs of readers who use them."76 From the reader's point of view, "functional documents," which the researchers define as documents which regulate readers' actions, must be structured around a function, such as to comply, to petition, and so on. Through protocol analysis,77 they learned how readers who read and paraphrased the meaning of functional documents structured the information in the documents. From the protocols emerged comments which did not pertain to the content or meaning of the regulation (meta statements) and three kinds of statements about content: structural, retrieval, and scenario. The revisions that emerged most frequently from the 31 readers' protocols were scenario statements, comments in which readers restructured the information. "By creating conditions, agents, and actions, they embed the meaning of a regulation in situations and contexts--they make meaning concrete enough to be functional for the reader."78 The purpose of Atlas's 1979 study, which concentrated on the process of writing, was to "develop an easy-to-use, objectively scoreable test that would tap a wide variety of writing behaviors" to investigate differences between expert-novice writers. Two aspects of Atlas's study are especially relevant to this study. First, an assumption that underlies both Atlas's and the present study is that "in order to make legitimate inferences about writing processes, we must first control for the subjects' real-world knowledge." Second, the writing task was to compose a business letter, since "the attitude of the letter's recipient could be carefully defined in advance, thus forcing the reader into a rhetorical stance, while the topic of the letter could be made narrow enough to allow the experimenter to brief all the subjects about it during the experiment."79 Atlas concluded that popular classroom approaches to helping writers focus attention on audience, like priming the students with questions, may have little effect on how writers address their audiences. A writer must not only be aware of his audience's needs; he must also "have procedures for translating that knowledge into action, and he must feel the need to apply those procedures."80 32 Atlas's use of a detailed questionnaire to focus his writers' attention on the relevant issues in the writing task demonstrates the value of knowledge-controlled testing. In Following Instructions, Rose and Cox report on their study of the complex processing required by conditionals, constructions that frequently appear in forms. (Example: If you are X, do Z.) Drawing from formal logic and cognitive psychology, the researchers conclude that performance in following instructions varies with the type of construction used and that whenever possible, negated or embedded .clauses should not be used in instructions. In Style: Ten Lessons in Clarity and Grace,81 Joseph Williams classifies, explains, and offers solutions to stylistic problems in nonacademic discourse and thus contributes to our understanding of style in government and business writing. The stylistic studies of Williams and Rosemary Hake also investigate discourse generated in business contexts. In "Non- Linguistic Linguistics and the Teaching of Style,"82 Williams criticizes text-centered, quantitative theories that consider only features relevant to the structures of sentences rather than to the discourse in which they' appear and which shapes the sentences. Williams and Hake challenged the popular notion that counting T-units83 provides an accurate measure of syntactic maturity. They found that the number of clauses in a T-unit or the number of words in a clause does not reflect the degree of maturity a reader perceives a sentence to have in the context for which it was writ 33 ten. Williams and Hake were also eager to go beyond earlier studies which used various grade levels and superior adult writers and examine data from non-superior adults "who write for the everyday purposes of commerce and industry, who write hasty and unedited but surely, in some sense, mature prose.1" 4 Although studies designed to find ways to improve readability of official communications are generally applauded, document designers like Veda Charrow caution rewriters that translating officialese into "plain English"85 is not a simple process. "Rewriting comprehensible prose," she warns in Let the Re-Writer Beware, "requires specialized knowledge about language and an awareness of a number of misconceptions and pitfalls."86 The situation that prompts Charrow's admonition is analogous to the one presented by the majority of business communication texts. While the needs of the reader have historically been a major consideration in writing effective letters and memoranda, the emphasis has been on how best to organize and format the material for efficient decision-making by the reader. And although business texts have also emphasized style, most frequently writers are merely urged to be concise, to save the reader's time. In fact, as the brief survey above suggests, studies of stylistic strategies used in discourse generated in business contexts for specific purposes and for specific readers are rare. 34 Purpose in Written Discourse Lee Odell and Dixie Goswami have also studied the nature and uses of writing in bureaucratic settings. In their reports on a three-year study funded by the National Institute of Research,87 they describe the research procedures that they developed to study nonacademic discourse. On the basis of their findings, they have warned researchers to be cautious, for instance, in making assumptions about the kinds of cohesion that will appear in different types of discourse. They found, for example, that some adult writers use lexical cohesion relatively frequently in one type of formal writing and relatively infrequently in one type of informal writing. At the 1982 Maryland Composition Conference, Odell described three procedures he and Goswami used to evaluate on-the-job writing. One procedure requires that evaluators collect samples of the full range of writing tasks performed by one writer. Then, they must ask the following question: "What kinds of choices reflect the writer's voice [persona], purpose, or relationship to his or her reader?"88 A second procedure is to observe the variations in style within the range of writing samples and interview the writer, asking two basic questions: "Here you do X. In other pieces of writing you do X or Z. In this passage, would you be willing to do Y rather than X? What basis do you have for preferring one alternative to the others?" 35 Finally, Goswami and Odell, using a procedure developed by Williams and Hake,89 created a piece of writing using the original data on which the writer had based his or her writing. They showed the writer what they had done and asked the writer to "Assume that I am a subordinate whom you have asked to do this writing. Would you be willing to use the draft I have written?" In brief, they found that the writing analysts relied heavily on rhetorical contexts to assess the pieces of writing. Several recent doctoral studies which have examined the relationship between purpose and stylistic features have an indirect bearing on the present discussion in that purpose of discourse is a component of each of the studies. In her doctoral study, "The Effect of Audience and Mode of Discourse on the Syntactic Complexity of the Writing of Sixth and Tenth Graders,"90 Maxine Crowhurst examined developmental differences in syntactic complexity in narratives, descriptions, and arguments (modes) written to two audiences differing in dimensions of age, intimacy, and power (best friend and teacher.) Compositions were analyzed for mean number of words per T-unit, mean number of words per clause, and mean number of clauses per T-unit. At grade ten, mode had a significant effect on all three measurements. At grade six, mode significantly affected the number of words per T-unit and the number of clauses per T-unit. 36 Grade 10--On W/TU, argument > description > narration On W/CL, description = argument > narration On CL/TU, argument > narration = description Grade 6--0n W/TU, argument > narration, argument = description description = narration On CL/TU, argument > narration = description Wilbur Pickering's study, "A Framework for Discourse Analysis," recognized the preference given to narrative discourse in studies and proposed "a framework for discourse analysis that will handle expository discourse in particular and will presumably be valid for any other discourse type as well as purpose."91 He concluded that the discourse system of a language and a text should be understood in terms of five organizing perspectives: hierarchy, cohesion, prominence, style, and strategy. Because they are discourse features that can be analyzed at the sentence and intersentence levels rather than at the discourse level, cohesion and style were the perspectives examined in the present study. Three recent doctoral studies examine purpose in business communications. In Ann Johns' comparison of cohesive elements in American business and non-native speaker written discourse, the non native speaker corpus consisted of samples of different rhetorical modes.92 Johns concluded that the mode of discourse employed was more important in determining incidence and distribution of cohesive elements than was the first language of the writer. Thus, she 37 recommends, ESL students would benefit from presentations which closely ally cohesive elements to mode of discourse. Harvey L. Addams determined written communication competencies most important to an accountant’s job effectiveness. In "Opinions of Practicing Accountants: Written Communication Competencies in the Accounting Profession,"93 he recommended that college accounting students be taught how to write narratives for audited and unaudited reports and comparisons and evaluations for analytic reports. Mildred S. Myers used rhetorical analyses of documents and interviews with managers and executives to show how managerial communication differs from professional or technical communication. In her study, "Written Communication at the Managerial and Professional/Technical Levels: A Case Study,"91* Myers looked at kinds of writing situations, discourse types, addressees, and rhetorical strategies characteristic of each group. The majority of managers wrote descriptive rather than analytic discourse; over half of the professionals* samples were analytic. One-third of the managerial level samples gave orders or instructions; professionals rarely wrote orders or instructions. Reader-Writer Relationship in Written Discourse To varying degrees, the doctoral studies of Donald L. Rubin, Crowhurst, and Myers consider how the relationship between the reader and writer affect writing style. In "The Development in Syntactic and Strategic Aspects of Audience Adaptation Skills in Written Persuasive Discourse,"95 for example, Rubin examined 38 developmental differences in stylistic and rhetorical strategies in relation to varied audiences (high, intermediate, and low intimacy) and a fixed purpose (persuasion). Persuasive writing from eighteen fourth-, eighth-, twelfth-grade and adult writers was analyzed. Syntactic complexity was measured by T-units; persuasive appeals were classified and counted; and social cognitive ability was measured in strategy rationales. Rubin found that longer clauses were directed to low intimacy readers and highly subordinated structures were characteristic of messages addressed to high intimacy readers. He found that writers view smaller-than-clause units as more intelligible and requiring less effort to process than subordinate clauses. In addition, Rubin found that writers felt obliged to maximize intelligibility for low intimacy readers, who had no prior motive to attend to the discourse. In Crowhurst*s study, sixth- and tenth-graders wrote narrations, descriptions, and arguments to two audiences, best friend and teacher. For all readers and all modes, grade ten compositions were more syntactically complex than the compositions written by the sixth-graders. In both grades, compositions written to teachers were more syntactically complex than those written to best friends. Among the rhetorical components examined by Myers was what audiences each group (managers/executives and professional/ technical employees) addressed most frequently. The relevance of Myers' study to the present one lies in the contrast between the 39 directions of the communications observed by Myers and the direc tions of communication in the memoranda discussed here. Decision Style What we know about how writers compose and how readers process prose has come largely from cognitive psychology. Using insights from cognitive psychology, Linda Flower and John Hayes, for example, have attempted to demystify the writing process.96 And our debt to cognitive psychology for demonstrating how language is acquired, processed, and stored, is made clear in comprehensive summaries like Eleanor J. Gibson and Harry Levin's The Psychology of Reading97 and bibliographies like Marshall Atlas's chapter in Document Design: A Review of the Relevant Research.98 Also from cognitive psychology has come an interest in matching writing style and information processing style with cognitive style. For example, Susan D. Kaufman's doctoral study, "Cognitive Style and Writing: An Inquiry,"99 examines the nature of the relationship between the individual cognitive styles of a community of freshmen and certain characteristics of their written products and composing processes. Writing samples were holistically scored and analytically scored for six criteria: content, organization of ideas, expression of ideas, usage and grammar, and mechanics and vocabulary. Kaufman found significant differences among certain characteristics of written products and responses to the attitude survey of writers in three cognitive style groups: descriptive- analytic, relational-contextual, and categorical-inferential. 40 In "A Study of the Linguistic Dimension of Information Processing as a Function of Cognitive Complexity," a highly problematic investigation, Gary F. Soldow "is concerned with the importance of tailoring the syntactical structure of a message to a particular receiver in light of the structure of the mind of that receiver."100 Soldow tested the hypothesis that people high in cognitive complexity (that is, the number of categories of struc tures in the mind that allow a person to process language) should be better able to process syntactically complex sentences than people low in cognitive complexity. In his study, cognitive complexity was measured by Crockett's Role Category Questionnaire, in which the number of descriptive adjectives that subjects use to describe a liked female, a disliked female, a liked male, and a disliked male is an index of their cognitive complexity. To determine how the subjects processed information, subjects were asked to give the meaning of a set of eight complex sentences that contained different combinations of marked and unmarked, right and left embeddings. He found that people high in cognitive complexity recalled more than people low in cognitive complexity and concluded that low complexity people "might not acquire important information that might expand or change the basic thought being expressed in a sentence."101 Soldow concludes, as do most business communication texts, that writers must adapt their prose to their readers’ needs. In addition to determining if a relationship exists between a writer's cognitive style and his writing style, researchers have 41 e also tried to establish a relationship between writing style and other variables, for instance between leadership style and writing style. In his doctoral study, "A Study of the Relationship Between Leadership Style and Writing Style for Supervisors and the Effects of Supervisory Experience, Formal Education and a Management Training Program on This Relationship,1,102 Larry R. Smeltzer examined the dimensions of consideration and structure in supervisor leadership style by administering the Leadership Opinion Questionnaire. Using content analysis to determine the writing style of two supervisors, Smeltzer found no significant relationship between leadership style and writing style in the consideration dimension. However, he found a negative relationship for the dimension of structure. Furthermore, no significant relationship was found between writing style and supervisory experience or between writing style and formal education. In addition, he found that management training did improve writing style. Another study, related to the present study because it examines writing samples from business and considers reader-writer relationships, is Tom J. McRorey's "The Role of Personality Characteristics in Written Business Communication: A Study of the Personality Characteristics of Public Relations Letter Writers in the Finance Industry in Tulsa, Oklahoma."103 Writers from fourteen public relations departments in banking composed responses to a common problem. The readers, thirty members of civic groups, evaluated the messages in terms of a semantic differential rating 42 scale. McRorey found that similarity of personality is not a sufficient predictor of reader response. As suggested in Chapter I and this summary of studies, the number of sources that address theoretical and practical concerns in nonacademic writing is increasing. Researchers have begun to use linguistic analyses and insights from cognitive psychology to describe readers' needs to suggest audience-oriented revisions. However, studies of nonacademic discourse more frequently raise questions than answer them. Questions about the role of purpose in shaping written discourse and about specific audiences and how to analyze their needs recur. Increased knowledge about nonacademic discourse has come from research in organizational behavior and management information systems. In studies of managerial decision styles, as this review shows, generalizations about the relationship between managers' decision styles and their writing styles have been made, but these generalizations have been based primarily on intuition. Until the present study, no one had examined systematically the relationship between writing style and decision style, a style that entails cognitive style and several other variables. 43 CHAPTER III METHODOLOGY Collection of Writing Samples A partner in the Los Angeles office of a major accounting firm solicited interoffice memoranda from management-level accountants in his firm. 104 (A copy of the,proposal submitted to the firm and the letter soliciting the cooperation of the accountants are in Appendixes A and B, respectively.) Following is a description of the collection procedure. 1. Twenty-three accountants completed Alan J. Rowe’s Decision Style Inventory (see Appendix C). The dominant and back-up decision styles of each accountant were identified as directive, analytic, conceptual, or behavioral. Rowe’s Inventory, which is described in Chapter 1, was chosen for this study for both theoretical and practical reasons: it is based on cognitive style and the rhetorical situation and can be administered speedily. 2. Each of the twenty-three accountants completed a fact sheet (see Appendix D) indicating his or her age, sex, years with the current firm, years in the accounting field, level to which most of his or her memoranda are directed within the firm, kinds of writing tasks (e.g., audit reports, engagement letters) most fre quently called for, an estimate of the average number of memoranda written per week, how much training in writing he or she had in 44 college and on the job. Answers to these questions were to be used to gather a writing sample that represents the spectrum of management-level accountants in the firm. The questionnaires indicated that three of the accountants did not write on the job. The remaining twenty accountants were asked to submit twelve to fifteen memoranda. (A copy of the letter to the accountants is in Appendix E.) 3. Of the twenty accountants from whom memoranda were requested, seven submitted a total of 67 memoranda. Those who did not provide memoranda reported that they communicated verbally and/or wrote only working papers, which I had excluded in my re quest. Distribution of memoranda among the dominant/back-up deci sion styles is shown in Table 1. (Summaries of the backgrounds of the seven accountants are in Appendix F.) 45 TABLE 1 NUMBER OF MEMORANDA ACCORDING TO DOMINANT AND BACK-UP DECISION STYLES Accountant Dominant Decision Style Back-Up Decision Style Number of Memos 1 Analytic Directive 1 2 Analytic Directive 16 3 Analytic Conceptual 12 4 Analytic Directive 7 5 Conceptual Directive 7 6 Conceptual Directive 11 7 Analytic Conceptual 13 Total 67 4. In the request for twelve to fifteen memoranda, the accountants were asked to identify the purposes of their memoranda using this list of speech acts: 1. To Describe (e.g., to describe a procedure, which may also entail classifying, identifying) 2. To Recommend (To propose, direct, order) 3. To Request (To ask, inquire) 4 . To Document (To verify) The accountants identified some memoranda as having two or more purposes. For those cases, Category 5 was created. (In Appendix G, 46 all of the purposes are illustrated.) 5. The accountants were also asked to identify the reader of each memorandum using this list: 1 . To staff above 2. To staff below 3. To staff at same level 4. To the file 5. To other (Please specify) The accountants indicated that some memoranda were directed to two or more readers. For those cases, Category 6 was created. Table 2 shows the directions in which each writer addressed his or her memoranda (see next page). Each memorandum was identified by the writer's number (1 through 7) and by numbering each one sequentially. Three copies of each were made, and the total was divided randomly into six packets with care taken to prevent duplicating memoranda in two packets. Identification of Memoranda Purposes Development of Memoranda Purpose Ratings To determine if memorandum style is affected by the purpose of the communication, it was necessary to identify the purpose of each memorandum. Because I was more concerned with illocutionary effect than with illocutionary intent, I used the purposes identified by 47 TABLE 2 MEMORANDA CLASSIFIED ACCORDING TO DIRECTION OF COMMUNICATION AS IDENTIFIED BY WRITERS Accountant (Number of Memoranda) Reader Above Reader Below Reader Same Level To File Outside Office Multiple Readers 1 (1) 101 (1,2) 2 (16) 213 214 203 204 208 3 201 202 205 206 207 209 210 212 215 216 211 (Client) 3 (12) 302 303 307 308 311 301 304 305 306 309 310 312 (2.3) (2.3) (2.3) (2.3) (2.3) (2.3) (2.3) 4 (17) 403 407 401 404 405 406 (Client) (Client) (Other Office) (Other Office) 402 (2,4) 5 (17) 502 503 504 505 501 506 507 (CPA Board) (Client) (Editor) 6 (11) 604 605 603 606 607 501 602 608 609 610 611 (1.3) (1.3) (1.4) (1.4) (1.4) (1.3.4) 7 (13) 705 701 703 709 713 704 706 708 712 702 707 710 711 (1.3) (1.3) (2.3) (2.3) Total Memos Per Level 9 5 21 5 8 19 48 three trained readers instead of those ascribed by the authors of the memoranda.10 5 First Reading of Memoranda; Three experienced college composition teachers identified the purposes of a set of memoranda using the same directions the accountants had followed (see Appendix H for a copy of the directions to the trained readers). These readers recommended that future readers would benefit from clearer distinctions between the categories of descriptions of procedures and documentations and recommendations and requests. Pilot Test of Rating Scale: To refine the set of purposes, nineteen experienced college composition teachers, including the three original readers, were divided into groups of six, depending upon the convenience of assembling six readers per session. They received the same written instructions as the first readers had followed. However, I discussed the differences between descriptions and documentations and between recommendations and requests. In the latter case, I asked them to assign "request" to memoranda that called for an answer, and "recommendation" to those that called for an action. After the discussions, each memorandum was assigned a purpose by six readers. Final Rating of Memoranda Six of the nineteen readers became the final arbiters of purpose in the memoranda. They were readers whose identification of purposes agreed with the majority of the other five readers on all memoranda they had read. 49 Before the six began the final reading, I used a modified version of the procedure described by Britton and his colleagues in The Development of Writing Abilities (11-18) and gave the readers a "briefing document." In the case of the Britton study, the document was written; the one used in this study consisted of having the readers evaluate and then discuss with me the main features of eight memoranda whose purposes had been unanimously agreed upon by six of the original readers and the authors. These eight memoranda served as models for the raters as they read and ascribed purposes to the memoranda. Care was taken to be sure no one re-read memoranda from the previous session. Rater Agreement When rater reliability was first calculated, using the method recommended for nominal scales by William A. Scott,106 it was found to be 0.38. This low rater reliability score was attributed to the fact that sixteen of forty-two memoranda on which there was two-thirds rater agreement had been identified as combinations of the same two purposes (Categories 1 and 4). Because raters had found a fifth category, rater reliability was recalculated to include Category 5. It was found to be 0.6*5. Table 3 shows the final purpose ratings assigned by the raters. 50 TABLE 3 PURPOSES OF MEMORANDA IDENTIFIED BY THREE TRAINED READERS A c c o u n t a n t s 1 2 3 4 5 1 6 7 Memo 1 4 4 1 2 2 3 2 2 2 1 1 4 4 4 4. 1 1 4 1 1 1 2 2 2 2 2 2 1 1 1 4 3 3 2 4 4 1 1 1 2 3 2 2 2 2 2 2 2 2 1 4 4 1 2 2 1 2 2 3 4 2 2 2 2 2 2 2 2 3 1 1 4 4 4 3 4 4 1 5 2 2 4 2 2 1 2 2 1 2 2 1 3 3 2 3 3 3 6 3 3 3 2 2 1 4 4 2 1 1 4 2 2 1 1 1 4 7 4 4 3 2 2 1 4 4 1 1 1 1 3 3 1 3 3 3 8 2 2 1 1 1 2 4 4 4 1 1 1 9 1 1 4 2 2 1 4 4 1 4 4 2 10 2 2 2 2 2 2 4 4 4 2 4 1 11 2 2 3 1 1 1 1 1 4 2 2 1 12 4 4 4 2 2 1 1 1 1 13 4 4 1 2 2 2 14 4 4 1 15 2 2 4 16 2 2 2 Coding Stylistic Features Four coding systems, described in Chapter 1, were used to isolate stylistic features in the memoranda: Prince, Frader, and Bosk's hedges; Halliday and Hasan's four categories of conjunctions; Christensen's sentence openers; and Halliday and Hasan's types of reference.107 In addition, modals, kinds and voices of verbs, grammatical subjects, and formulary expressions were isolated and coded. Table 4 shows the features in each category (see next page). A sample coded memorandum appears in Appendix I. 51 TABLE 4 STYLISTIC FEATURES CODED Hedges Adaptors Rounders Plausibility Shields Attribution Shields Conjunctions Additives Adversatives Causals Temporals Grammatical Subjects You understood Pronouns (Excluding Personal) Names of People (Including Personal Pronouns) Nominals Others References Personal Pronouns Possessive Determiners Demonstrative Pronouns The Modals Can/Could May/Might Must Should/Ought to Will/Would Negative Others Sentence Openers Clauses Prepositions Adverbs Nouns in Adverb Function Present Participles Past Participles Infinitives/Gerunds Inverted Passive Verbs Inverted Progressive Verbs Kinds and Voices of Verbs Passive Voice Verbs Active Voice Verbs Linking Verbs Intransitive Verbs Indicative Verbs Formulary Expressions Future Contact Acknowledge Request Solicitous Comment 52 Statistical Analysis Information about each stylistic feature was summarized by keypunching the data and analyzing it using the subprogram Frequency from the Statistical Package for the Social Sciences (SPSS). Obtained from this procedure were frequency counts of each stylistic feature in the total writing sample, plus frequency counts for subsets of the memoranda. These counts were selected by the deci sion styles of the writers of the memoranda, the purposes of the memoranda, and the intended readers as identified by the accountants who wrote the memoranda. 53 CHAPTER IV DECISION STYLE AND WRITING STYLE Decision style researchers like Driver and Rowe have suggested that writers' decision style preferences influence their writing styles. To study the extent to which the generalizations hold, I tabulated the frequencies with which selected stylistic features appear in memoranda written by seven accountants in a major national accounting firm. Their dominant and back-up decision styles were identified through Rowe's Decision Style Inventory as Analytic- Directive, Analytic-Conceptual, Conceptual-Directive, and Conceptual-Analytic. The linguistic features selected to identify writing styles were chosen on the basis of findings in discourse analysis and sociolinguistic studies: linguistic elements that have been shown to help create cohesive and coherent discourse and to imply relationships between writers and their intended readers. Analyses of the memoranda suggest that writers' choices of cohesive devices, elements that create coherence, and features that imply reader-writer relationships are not greatly influenced by the writers' decision styles. There are, however, some notable exceptions to that generalization. Because both managerial decision style theorists and literary stylists contend that common traits are exhibited by distinct groups, it is somewhat surprising to observe that most linguistic 54 features remain fairly consistent across the four decision styles and that few of these features can be associated with one style. The consistency with which groups of decision stylists employed the linguistic devices selected for identifying writing style is reflected in the figures in Tables 5, 6, and 7. Following each table are discussions of the several exceptions to the consistency among the decision stylists' choices. 55 Cohesive Elements TABLE 5 DECISION STYLE AND COHESIVE ELEMENTS: FREQUENCIES OF COHESIVE ELEMENTS IN MEMORANDA IN T-UNITS (T-U) (T-Units Per Category Shown in Parentheses) Analytic- Directive (90 T-U) Analytic- Conceptual (376 T-U) Conceptual- Directive (196 T-U) Conceptual Analytic (133 T-U) Reference Items Personal Pronouns 1.13 ' .25 .35 .20 Possessive Determiners .23 .16 .31 .15 The .98 .95 .96 .71 Demonstratives . 14 . 13 .13 . 11 Total 2.48 1.49 1.75 1.17 Grammatical Subiects You Understood . 18 .26 .16 .05 Pronouns .06 .05 .03 .02 People .44 -15 .32 .12 NominaIs .22 .29 .25 .53 Others . 17 .22 .28 .27 Total 1.07 .97 1.04 .98 Verbs and Voices Passive Voice .21 .27 .27 .39 Linking . 19 .16 .32 •I4 Active Voice- .42 .26 .40 .37 Intransitive .07 .04 .06 .02 You Understand . 18 .26 . 16 .05 Total 1.07 .99 1.21 .97 As one would expect, the is the most frequently used reference item across decision styles. And among the stylists, Analytical- 56 Directive decision stylists use more reference items than other stylists use. Analytic-Directive stylists' use of the nearly equals other stylists' uses, but they seem to prefer pronouns over the, suggesting that they write messages that discuss people and peoples' actions. In the memorandum reproduced below, those stylistic characteristics are apparent. Memorandum 212. Analytic-Directive Subject: [Deleted in Original] Enclosed are all working papers related to the procedures you requested we perform at XYZ. I understand that X has told you that our audit at XYZ for the year ended January 31, 1982 is not yet complete. At this time, we believe there will be an adjustment related to accounts receivable that may affect the December 31, 1981 balances. Also, we have made an adjustment of approximately $1,200,000, reducing inventory and accounts payable at January 31, 1982. The client has not yet provided the data needed to determine if this entry affects the December 31, 1981 balances. We will inform you of any changes to the December 31, 1981 balances as soon as the information is available. As the figures in Table 5 indicate, nearly 60 percent of the grammatical subjects used by Analytic-Directive stylists refer to people. This preference suggests that the writer presumes the reader shares with him or her knowledge of the referents, either from the text or from outside the test. Furthermore, many of those people subjects are "you understood." In contrast, Conceptual- Analytic stylists rarely use people as subject. Instead, half of their grammatical subjects are nominals, a preference that suggests 57 attention to activities (departure, examination, assistance) and concepts (responsibilities). When one combines the active voice verb counts with the “you understood" subject counts, it is clear that the Agent-Action-Goal order is preferred in over half of the sentences written by all stylists except those of Conceptual-Analytic stylists. They prefer passive voice verbs, using passives in two-fifths of their main clauses. (A main clause, or independent clause, consists of a subject, or coordinated subjects, with a finite verb, or coordinate verbs.) The following memorandum illustrates the Conceptual-Analytic stylists’ preferences for passive voice verbs (parentheses indicate active voice verbs): Memorandum 605. Conceptual-Analytic Subject: [Deleted in Original] Enclosed are the following workpapers relating to our December 31, 1980 examination of [Deleted in Original]. • General Binder • All Workpapers - Los Angeles • All Workpapers - New York As noted on W/P Z-2 of the general binder, the report has not been issued as the client was unable to provide evidence to support collectibility of affiliate receivables. Please (let) me (know) when to schedule the December 31, 1981 examination of the Los Angeles office. Tax files will be forwarded separately by Mike Davis of the tax department. Please (acknowledge) receipt of these workpapers on the attached copy of this memorandum. 58 Elements of Goherence As indicated by the figures in Table 6, words or phrases, i.e., sentence openers, precede grammatical subjects in 4l percent of the sentences written by Analytic-Directive stylists. This figure represents a high incidence of sentence openers compared to the 23 and 24 percent that appear in Analytic-Conceptual and Conceptual- Directive stylists' writing and the 18 percent in Conceptual- Analytic stylists'. TABLE 6 DECISION STYLE AND ELEMENTS OF COHERENCE: FREQUENCIES OF ELEMENTS OF COHERENCE IN MEMORANDA IN T-Units (T-U) (T-.Units. Per Category Shown in Parentheses) Analytic- Directive (90 T-U) Analytic- Conceptual (376 T-U) Conceptual- Directive (196 T-U) Conceptual Analytic (133 T-U) Con iunctions Additives .11 .06 .05 .02 Adversatives .04 .01 .04 .00 Causals .04 .03 .01 .05 Temporals .01 .01 .03 .00 Total .20 .11 .13 .07 Sentence Openers Adverbials .30 .20 .20 .11 Verbals .03 .01 .03 .04 Inverted Constructions .08 .02 .01 .03 Total .23 .24 • .18 Percentages of sentence openers in Christensen's 4000-sentence corpus selected from the beginnings of narrative and discursive 59 prose written by twenty professional writers are not, or course, directly comparable to the percentages of openers in the 795 T-units of complete pieces of discourse examined in this study. However, it is interesting to note that while about 25 percent of his writers placed something before their subjects, Analytic-Directive and Conceptual-Directive stylists began 40 and 30 percent of the T-units in their transactional discourse with sentence openers. And while 23 percent of Christensen’s openers were adverbials, Analytic-. Directive stylists opened 30 percent of their sentences with adverbials, and Conceptual-Analytic stylists used only 11 percent adverbials. Conjunctions, however, are used infrequently across decision styles. It may be that in the accounting memoranda, sentence openers partially replace conjunctions in connecting propositions and therefore account for the scarcity of conjunctions. For example, in the following memorandum, "As we have * previously discussed" connects the proposition "we have enclosed the trial balance" with "we have documentation and recordkeeping problems." "Based on the results of the work done in Compton," connects "the working papers support the inclusion of the Compton trial balance in the X financial statement" with "you requested work in your February 8, 1982 memo." 60 Memorandum 209. Analytic-Directive Subject: [Deleted in Original] Enclosed is the trial balance for [X] (after making final adjustments on May 13, 1982) for inclusion in the [X] limited financial statements. As we have previously discussed, due to documentation and recordkeeping problems at Compton, we had to significantly extend our procedures to complete the work that you requested in your memo dated February 8, 1982. Based on the results of the work done at Compton, we are satisfied that the working papers support the inclusion of the Compton trial balance in the [X] financial statements; however, as we discussed, our approach was largely balance sheet-oriented and the records did not provide adequate information to support reporting separately on Compton’s statement of operations. If you have any questions, please call. Features That Imply Reader-Writer Relationships As the figures in Table 7 suggest, uses of hedges and modals, linguistic features that generally reflect writers' attitudes toward their readers, are apparently not influenced by writers' decision styles. Furthermore, perhaps the low incidence of hedges and modals in interoffice memoranda can be attributed to the fact that the concern for maintaining goodwill with the reader is not as great as it is in correspondence directed outside the office. 61 TABLE 7 DECISION STYLE AND DIRECTION OF COMMUNICATION: FREQUENCIES OF READER-WRITER RELATIONSHIP ITEMS IN MEMORANDA IN T-UNITS (T-U) (T-Units Per Category Shown in Parentheses) I / Analytic- Directive (90 T-U) Analytic- Conceptual (376 T-U) Conceptual- Directive (196 T-U) Conceptual Analytic (133 T-U) Hedges Approximator/Adaptors .01 .02 . -05 .02 Approximator/Rounders .07 .03 .06 .05 Plausibility/Shields .09 .02 .04 .02 Attribution/Shields ■i .01 .01 .02 .03 Total . 18 .08 .17 . 12 Modals Can/Could .01 .01 .02 .01 May/Might .00 .01 .02 .01 Must .01 .01 .01 .00 Other .01 .03 .01 .01 Negative .07 .06 .08 . 10 Should/Ought To .04 .08 .08 .05 Will/Would . 19 .11 .05 .21 Total .33 .31 .27 .39 Personal Pronouns First-person Singular .46 . 10 . 10 .03 First-person Plural .'27 .09 .30 .20 Second-person Pronouns .63 .37 .34 .10 Third-person Pronouns .29 .06 .09 . 12 Total 1.65 .62 .83 .45 Formulary Expressions Future Contact .04 .01 .02 .00 Request Acknowledgment .02 .01 .01 .02 Solicitous Comment .01 .01 .01 .00 Total .07 .03 .04 .02 62 Frequencies and kinds of pronouns used in memoranda do, however, appear to be influenced by writers' decision styles (see Table 7). Analytic-Directive stylists' preference for first person pronouns over second person pronouns is greater than others1 preferences. Of greater significance, however, is the fact that Analytic-Directive stylists use 1.5 pronouns per T-unit, contrasted to the 0.2, 0.9, and 0.4 per T-unit employed by Analytic-Conceptual, Conceptual-Directive, and Conceptual-Analytic stylists, respec tively. This preference suggests that their memoranda are people- oriented. Furthermore, their preferences for first person over second person pronouns suggest that their memoranda describe company activities and, more than likely, because of the heavy use of first person singular, the writers' own activities. The following excerpt illustrates the pronoun preferences of Analytic-Directive stylists: Memorandum 215. Analytic-Directive Subject: [Deleted in Original] Enclosed are the three working paper binders related to the June 30, 1981 audit work done in Atlanta. _I understand that they are needed for a planning meeting with the client. j[ would appreciate the return of these binders to Los Angeles when the present need is completed so we can utilize them in our planning of the overall audit for 1982. Please acknowledge receipt by signing and returning the enclosed copy of this memo. The accountants closed twenty-two memoranda, nearly one-third of the writing sample, with formulary expressions that either 63 solicited further communication from readers, requested acknowledgments of the writers' memoranda or materials referred to in them, or made solicitous comments. Analytic-Directive and Analytic-Conceptual stylists conclude nearly a third of their memoranda with expressions like these: If you need any additional information, please call. Please acknowledge receipt by signing and returning the enclosed copy of this memo. Thank you for your cooperation. Conceptual-Analytic stylists, on the other hand, conclude only two of their eleven memoranda with formulary expressions, and Conceptual-Directive stylists use formulary expressions in six of their fourteen memoranda. From these figures, we can only hypothesize a relationship between decision ‘style preference and use of formulary expressions. Conclusion Decision style appears to influence stylistic choices most frequently in the writing of Analytic-Directive decision stylists: They use more reference items than others, specifically preferring personal pronouns over other reference items isolated in the study and using them more frequently than other stylists do. In addition, they use sentence openers more frequently than others, using adverbials, verbals, and inverted constructions about 40 percent of the time. 64 From these few generalizations about stylistic preferences of one decision style group, we cannot conclude that decision style preferences influence writing styles. However, because the writing of Analytic-Directive stylists differs from other stylists' writing so dramatically in the few features discussed above, we can hypothesize a relationship between Analytic-Directive stylists and their linguistic choices. Furthermore, we cannot ignore the possibility that features other than those isolated here may support claims that a relationship indeed exists. 65 CHAPTER V PURPOSE AND WRITING STYLE One assumption in rhetorical theory is that a writer's purpose shapes his or her stylistic choices. Recently, that assumption has been validated in the doctoral studies of Crowhurst and Meyers and in the three-year study of nonacademic,discourse reported by Lee Odell and Dixie Goswami. To determine if purpose is a controlling factor in accountants’ linguistic choices, I tabulated the frequencies of selected stylistic features in memoranda which trained raters had identified as procedures, recommendations, requests, and documentations,. or a combination of description and documentation. Analyses of the memoranda offer further evidence that style in discourse is in part influenced by the purpose of the discourse. Results of the analyses of cohesive devices, elements that create coherence and features that imply reader-writer relationships appear in Tables 8, 9, and 10. Cohesive Elements Reference items seem to be used with nearly the same frequency across three of the five categories, with requests and recommendations employing more reference items. Choices of one 66 reference item, personal pronouns, appear to be influenced by discourse purpose. TABLE 8 PURPOSE AND COHESIVE ELEMENTS: FREQUENCIES OF COHESIVE ELEMENTS IN MEMORANDA IN T-UNITS (T-U) (T-Units Per Category Shown in Parentheses) Description of Procedure (349 T-U) Recommendation (204 T-U) Request (33 T-U) Documentat ion (195 T-U) Description/ Document at ion (301 T-U) Reference Items Personal Pronouns .19 .51 1.00 .39 .28 Possessive Determiners . 16 .29 .30 .19 . 19 The .91 1.01 .88 .81 .88 Demonstratives .09 . 18 .06 . 15 . 10 Total 1.35 1.99 2.24 1.54. 1.45 Grammatical Subjects You Understood .23 .25 .22 .01 .03 Pronouns .04 .06 .03 .02 .05 People .14 .25 .42 .29 .27 Nominals .33 .30 . 18 .33 .41 Others .24 . 18 . 15 .32 .26 Total .98 1.04 1.00 .97 1.02 Verbs and Voices Passive Voice .28 .30 .24 .28 .26 Linking . 11 . 18 .15 .24 .26 Active Voice .35 .28 .39 .33 .38 Intransitives .04 ( .03 .06 .06 .07 You Understood .23 .25 .21 .07 . 03 Total 1.01 1.04 1.05 .98 . 1.00 67 Personal pronouns appear five times more frequently in requests than in descriptions of procedures, twice as frequently as in recommendations. Writers of the six memoranda identified as requests asked their readers for replies, as these excerpts illustrate: If you are aware of anyone, within XYZ or elsewhere, who meets the qualifications below and is interested, please call me at once. [You] Please call me if you have any comments or questions. We will await your approval prior to issuing the letter to local management. [You] Please let me know when to schedule the December *31, 1981, examination of the Los Angeles office. Descriptions of procedures, on the other hand, contain the fewest personal pronouns, since they more frequently refer to companies, activities, reviews, and inventories than to people. Choices of grammatical subjects are also affected by purpose in several ways that one would expect. "You understood" subjects are rare in memoranda written to document conversations and past and future activities and to describe/document. People appear most frequently as subjects in memoranda that request replies and the least in descriptions of procedures. Nominals appear most often in memoranda that describe/document and the least in requests. Although passive voice and intransitive verbs are used in nearly equal proportions among the five categories, there are differences among the other kinds of verbs and their frequencies in the various categories of purpose. Verb choices in documentations, 68 for instance, differ noticeably from those in other categories. Linking verbs make up a fourth of all of the verbs in documentation and description/documentation memoranda, whereas they represent only 10 percent of the verbs in descriptions of procedures. Furthermore, command sentences, it is no surprise, are rare among documentations and description/documentations; these are memoranda that require no responses or reactions. (Memorandum 503 on page 72 illustrates the kinds and frequencies of verbs in documentation memoranda.) Elements of Coherence Although uses of conjunctions are not affected by purpose, uses of sentence openers may be influenced by the purpose of memoranda. 69 TABLE 9 PURPOSE AND ELEMENTS OF COHERENCE: FREQUENCIES OF ELEMENTS OF COHERENCE IN MEMORANDA IN T-UNITS (T-U) (T-Units Per Category Shown in Parentheses) Description of Procedure (349 T-U) Recomraendat ion (204 T-U) Request (33 T-U) Documentation (195 T-U) Description/ Documentation (301 T-U) Con iunctions Additives .06 .07 .09 .03 .03 Adversatives .02 .02 .03 .02 .02 Causals .04 .05 .00 .01 .02 Temporals .01 .01 .00 .02 .00 Total .13 . 15 . 12 .08 .07 Sentence Openers Adverbials . 18 .27 .24 . 14 .16 Verbals .02 .02 .00 .02 .03 Inverted Construct ion: .01 .03 .09 .02 .01 Total .21 .32 .33 . 18 .20 For instance, a third of the grammatical subjects in recommendations and requests are preceded by words, phrases, or clauses. Many of these offer reasons for the recommendations and requests, as these excerpts illustrate: If made, I feel that these changes could further enhance the program and reduce some misunderstanding that developed during my period as a [X] . Even though the very nature of the program does not easily lend itself to structure, I feel that if we can develop some loose framework, everyone involved will benefit. 70 Accountants who wrote requests and recommendations used more sentence openers (33 percent) than Christensen's professional writers used (25 percent), a difference due, in part, to the narrative and discursive purposes of the passages Christensen analyzed. Features That Imply Reader-Writer Relationships In general, accountants used few hedges in their memoranda, as the figures in Table 10 illustrate. However, there does appear to be a relationship between uses of hedges and purpose of discourse. TABLE 10 PURPOSE AND DIRECTION OF COMMUNICATION: FREQUENCIES OF READER-WRITER RELATIONSHIP ITEMS IN MEMORANDA IN T-UNITS (T-U) (T-Units Per Category Shown in Parentheses) Description of Procedure (349 T-U) Recommendatloi (204 T-U) Request (33 T-U) Document at ion (195 T-U) Description/' Documentaticr (301 T-U) Hedges App roximator/ Adaptors .02 ,02 .03 .05 .05 Approximator/ Rounders .02 .06 .03 .06 .04 Plausibility/ Shields .01 .03 .18 .05 .02 Attribution/ Shields .01 ■ 01 .06 .02 ■ 01 Total .06 . 12 .30 . 18 .12 Modsla Can/Could .01 .01 .00 .02 .01 May/Might .02 .00 .00 .01 .01 Must .00 .02 .00 .01 .00 Other .01 .05 .00 .02 .01 Negative .07 .04 .06 .09 .08 Should/Ought To .03 .19 .00 .03 .02 Will/Would ■ 15 ■ 12 .12 ■ 08 .13 Total .29 .43 .18 .26 .26 Personal Pronouns First-person Singular .05 .22 .52 .12 .09 First-person Plural- .15 . 16 .18 .16 .18 Second -person Pronouns .28 .53 .48 .15 .06 Third-person Pronouns .08 . 15 .30 .18 .15 Total .56 1.06 1.48 .61 .48 Formulary Expressions Future Contact .01 .03 .03 .01 .01 Request Acknow1edgment .00 .02 .03 .02 .00 Solicitous Comment .00 .00 .00 .00 .00 Total .01 .05 .06 .03 .01 72 Nearly one-third of the T-units in the six requests have hedges, as contrasted to descriptions of procedures, which have hedges in only 6 percent of the T-units. Moreover, the most common hedges in requests are Plausibility Shields, examples of which are underlined in the memorandum below: Memorandum 503. Request. Subject: [Deleted in Original] I recently attended the Annual Convention of [XXX] in Atlanta. While there I had the opportunity to have dinner with Mr. Y. Y indicated that they were considering putting the Audit out for bids. Apparently they are unhappy with the quality of service from their current auditors. [X] I thought that this may be of interest to our Z office since I don't know anyone in Z. I. thought that you could inform the appropriate people. If you have any questions, please feel free to call on me. Further generalizations about writers' preferences for hedges in kinds of discourse based on six memoranda would be at best speculative; however, one may find the appearance of .18 hedges per T-unit in documentations surprising, since those memoranda verify qualifications, report the status of engagements, and recount conversations, writing tasks that would seemingly elicit few Approximators or Shields from the writers. Nevertheless, as these examples from Memorandum 212, page 56, illustrate, hedges are appropriate to documentations: I understand that X has told you that our audit at XYZ for the year ended January 31, 1982 is not yet complete. (Plausibility Shield) 73 At this time, we believe there will be an adjustment related to accounts receivable that may affect the December 31, 1981 balances. (Plausibility Shield) Also, we have made an adjustment of approximately $1,2000,000 .... (Rounder) We will inform you of any changes to the December 31, 1981 balances as soon as the information is available. (Rounder) Approximately one-third of the T-units in the five categories of purposes contain modals. They are used most frequently in recommendations (.43 per T-unit) and least frequently in requests and description/documentations (.12 per T-unit). Writers of recommendations, as one might expect, use .21 must, should ought to's per T-unit. Additional phrases, such as are to be reviewed, is necessary, and is mandatory, add to the force of the recommendations in general. As the underlined phrases in the memorandum below illustrate, the force of the recommendations is usually explicit. Memorandum 310. Recommendation. Subject: Completion of Outstanding P-66's Performance evaluation forms (P-66's) should be prepared for all staff personnel on engagements completed as of October 24, 1980, and for personnel who have accumulated at least 40 hours on an engagement as of that date. The completed P-66's should be forwarded to the engagement manager no later than Monday, November 3, 1980. Your prompt attention to this matter is necessary to ensure proper and timely completion of the semi-annual reviews and career counseling sessions that are scheduled to begin on or about November 10, 1980. 74 The . 12 per T-unit use of modals in requests and description/ documentations contrast to the .26 average used across the purposes. This phenomenon may be accounted for in part by the fact that requests are frequently tempered by hedges and "if" clauses. In the case of description/documentations, one would expect few modals in documents that do not require the maintenance of goodwill or the implication of a manager-subordinant deference. Purpose affects the frequency with which personal pronouns are used in the memoranda. Overall, more pronouns are used in requests (1.48 per T-unit) than in memoranda written for other purposes, and fewer pronouns appear in description/documentation memoranda (^48). Moreover, purpose seems to affect the proportion of first-person to second-person pronouns. Second-person pronouns, for instance, appear more frequently than first-person pronouns in descriptions of procedures (.28 per T-unit to .20 per T-unit) and recommendations (.53 per T-unit to .38 per T-unit), where the writer must tell the reader what he or she needs to do in relation to the procedure or recommendation. In the case of recommendations, the second-person pronouns are generally nearly equally divided between "you understood" subjects and forms of "you" explicit in the text. The proportion of pronouns in recommendations is typified in Memorandum 210 below: 75 Memorandum 210. Recommendation. Subject: [Deleted in Original] Enclosed are the three working paper binders from the 1981 examination. [You] Please charge all time related to the 1982 examination at 80% of standard plus expenses to contract [Deleted in Original]. Also, [you] please include jne as an information addressee, Social Security #153-42-9536, when you open the contract. The additional information that we discussed today such as audit gauge and general audit plan, will be forwarded to you in the near future. [You] Please acknowledge receipt by signing and returning the copy of this memo. Writers make another distinction in their uses of personal pronouns: as the figures in Table 10 illustrate, purpose also seems to affect writers' choices of singular and plural first person pronouns. For example,, in descriptions of procedures, which may explain, for instance, the firm's inventory procedure, "we" is used. But when they recommend and request, they prefer using first-person singular. Finally, although there are only ninety-three formulary expressions in the 795 T-units, there may be a pattern in the writers' uses of the expressions. Since recommendations and re quests require reactions and responses, the appearance of expres sions about future contacts would be expected. Conclusions There are, then, patterns in writers' stylistic choices that can be seen in memoranda classified by purpose. In memoranda that 76 describe procedures, the accountants used the fewest linking verbs, "people" subjects, possessive determiners, and personal pronouns. Descriptions also contain the fewest hedges. On the other hand, they contain more second-person pronouns than first-person, and, in general, the fewest pronouns. In memoranda that make recommendations, accountants used the most command sentences, yet the fewest active verbs. They also used the most passive voice verbs, personal pronouns, demonstratives, and the. Causals appear most frequently in recommendations, as do sentence openers, especially adverbials. In recommendations, the writers used the most modals, more second- than first-person pronouns, and, along with requests, six times more formulary expressions than description and description/documentation memoranda, and twice as many as documentations. Memoranda that request have the most active verbs and the fewest passive voice verbs. They use the most "people" subjects, account for the most combined "you understood" and people, and the fewest nominals and "others." Writers of requests used the most personal pronouns and the least demonstratives. Additives and adversatives appear most frequently in them. Recommendations and requests combined contain two-thirds of the sentence openers. Nearly one-third have hedges and they have the fewest modals. In terms of pronouns, requests have the most first-person pronouns, the most singular first-person pronouns, and use more third-person pronouns than any other purpose. Like recommendations, requests use 77 six times more formulary expressions than descriptions and description/documentations and twice as many as documentations. Documentation memoranda have the most "other" and the fewest "you understood" and pronoun subjects. Writers of documentations use the more than other writers do. They also prefer first-person pronouns to second-person pronouns, 2 to 1. Description/documentations contain the most linking and intransitive verbs, the fewest command sentences, the most nominal subjects, and the most reference items. In addition, they use more first-person pronouns than second-person (4 to 1). 78 CHAPTER VI DIRECTION OF COMMUNICATION AND WRITING STYLE Discourse analysts like Crowhurst and Rubin and composition researchers like Atlas and Flower and Hayes have shown that writers make stylistic choices on the basis of who their intended readers are. More recently, in their three-year study of stylistic choices made by nonprofessional writers on the job, Lee Odell and Dixie Goswami found that linguistic choices like cohesive devices and pronouns are influenced by the writers' perceptions of their read ers . To determine if intended readers .of accountants' memoranda influence writing styles of accountants, I tabulated the frequencies with which selected stylistic features appear in memoranda whose readers had been identified by the accountants as above them, below them, and at their same level in the firm. Two other specified groups were "To the File" and "To Others." The accountants iden tified a sixth group— multiple readers— in 19 of the 67 memoranda. Analyses of accountants' memoranda confirm that, in general, direction of communication affects writers' choices of cohesive devices, elements that create coherence, and features that imply reader-writer relationships. Results of the analyses are shown in Tables 11, 12, and 13. 79 Cohesive Elements Reference items appear most frequently in communications to readers outside the accountants' office because writers must provide con texts for the information in documents. In the eight communications directed to outside readers, writers explain internal controls and readers' responsibilities in upcoming inventories and transmit documents that are mentioned in the memoranda, as in these sen tences : Please call me after you have reviewed the attached. Attached is a summary of our analysis of X. Enclosed are the revised draft financial statements for X. 80 TABLE U DIRECTION OF COMMUNICATION AND COHESIVE ELEMENTS: FREQUENCIES OF COHESIVE ELEMENTS IN MEMORANDA IN T-UNITS (T-U) (T-Units Per Category Shown in Parentheses) Reader Above (57 T-U) Reader Below (60 T-U) Reader at Same Level (153 T-U) To File (220 T-U Reader Outside (65 T-U) Multiple Readers (240 T-U) Reference Personal Pronouns .61 .53 .59 .07 .60 .34 Possessive Determiners .35 .08 .31 . 10 .37 .21 The .73 .83 1.00 .94 1.10 .86 Demonstrat ives .16 . 12 . 18 .09 . 15 .12 Total 1.85 1.56 2.08 1.20 2.22 1.5.3 Grammatical Subjects You Understood . 18 .50 .28 .20 .22 .05 Pronouns .05 . 10 .03 .04 .02 .03 People .53 . 12 .27 . 11 .34 .22 Nominals .18 . 17 .27 .33 .29 .40 Others .18 . 12 . 21 .27 .18 .28 Total 1.12 1.01 1.05 .95 1.05 .98 Verbs and Voices Passive Voice .18 . 17 .30 .27 .29 .34 ' Linking .32 .12 .17 .15 .05 . 18 Active Voice .29 .17 .27 .35 .45 .38 Intransitive .02 .03 .03 .05 .02 .06 You Understood . 18 .50 .28 .20 .22 .05 Total .99 .99 1.05 1.02 1.03 1.01 As the figures in Table 11 illustrate, the 21 memoranda written to readers at the same level as the writer contain nearly as many reference items as those written to outside-the-office readers. 81 This comparable use of reference items can largely be attributed to the fact that 15 of the 21 memoranda were addressed to outside readers; that is, the writers categorized the outside-the-office readers as their peers rather than as outside-the-office readers. A letter addressed to a newspaper editor is a reaction to a published discussion of interest rates. Another addressed to the CPA documents the writer’s work experience. Because these communications discuss attachments and summarize work experience, the and demonstratives appear frequently, and pronouns and possessives referring to the firm and to the reader are used. The influence of outside-the-firm readers on writers' choices of reference items is illustrated in the following memorandum. Memorandum 506. Outside-the-Firm Reader Dear X: Attached is a summary of our analysis of [Deleted in Original] Savings & Loan’s Now Account Item Processing Department. All FHLB cost figures represent cost quotes associated with a similar engagement one month prior to our study at [Deleted in Original]. Of course these costs and details of such a conversion will have to again be negotiated with the FHLB or similar vendor. The FHLB was used for analysis purposes because they provide a full line of inclearing services. Other vendors in the Portland area do not provide the same scope of service and require portions of the processing to be maintained inhouse (i.e. bulk filing, statement preparations etc.). This analysis was completed with a limited amount of data, and warrants further investigation. If ^ can be of any further assistance, please don't hesitate to call. Reference items appear least frequently in the five memoranda directed to the file and in the six memoranda for which writers 82 identified one of the multiple directions of the memoranda to be the file. These transactional communications are working papers for engagements, summaries of working papers, and documentations of conversations with clients. Repeated references to conditions within the readers' firms leads to the recurrence of the in nearly all T-units (.9). The following memorandum reflects approximately the proportion of kinds of reference items in memoranda directed to the file. Memorandum 708. To the file. Subject: DEF--1981 EPD Review As part of the analysis of internal control of Interna tional Customs Services (ICS) for the year ended December 31, 1981, a review of the EDP function was performed in accordance with Section 6000 of the Audit Manual. Work paper 507 (Client EDP Profile), as well as narrative descriptions and memoranda, can be found in the work papers. This memorandum serves to summarize and highlight the findings, and is based upon my discussions with A, B, and C, audit senior, as well as my observations during the review. Overview ICS utilizes a Basic Four Model 730 mainframe with 192K storage capacity at its Y headquarters. Its function is to. summarize data transmitted from remote branches and to perform certain corporate financial functions. Seven remote branches utilize either Basic Four Models 200 or 410 to process orders and invoice customers. Upon daily command from Y, the remote computers electronically transmit summaries of their daily activity to Y via telephone hookups. Each remote computer operates in a similar fashion using standard software developed by the corporate EDP depart ment. None of the regions employs personnel in EDP functions other than for data input/output functions. 83 Reviews of the remote EDP sites is [sic] not considered necessary due to certain manual controls discussed below. Involvement of a Senior Computer Audit Specialist is similarly not considered necessary. General Control The DP department is very small. EDP operators run the computer, have access to all programs and data files and make program changes. These environmental control weak nesses have precluded our reliance. The presence of these weaknessses means that primary assurance that data is processed completely and accurately must rest upon user controls and substantive audit procedures. Specific Applications A limited specific application review was performed to provide an understanding of the flow of transactions and to provide assurance that no pervasive weaknesses exist. This review reveals that the Company depends primarily on user controls (has totals, batch totals, control logs, and reconciliation procedures) to provide assurance of input, processing, and output. Possible Management Letter Comments Many of the traditional controls associated with data processing are not possible due to the small size and limited personnel of the client's DP operation. • Backup master and transaction summary files are stored on site for up to 30 days before being transferred to an outside location. This exposes the Company to certain file reconstruction problems. • Remote terminal access is based upon passwords coded within the application software, which are not changed. This exposes the Company to unauthorized EDP access which could be mitigated by occasional password changes. • Priorities for new data processing applications and enhancements of existing applications are not for malized and authorized within an EDP Users Committee framework. This exposes the Company to inefficient operations and underutilization of the EDP resource. Programming documentation is stored in open files accessible to many individuals without a need for the 84 information. This exposes the Company to unau thorized program changes. Conclusion Because of the lack of adequate segregation of duties within the DP department resulting from the limited number of the personnel and other reasons, the system of general controls at present is not conducive to reliance. _I agree with the audit team's decision to test the significant DP applications on a substantive basis without placing reliance on DP controls. Audit procedures should be appropriately designed to substantively test the audit implications of the weaknesses noted in the DP controls without placing reliance on DP controls. In the sales order system, this should include testing of the account ing department's 100% reconciliation of remote site EDP reports to copies of the transaction documents separately sent by the remote branches. Although the direction of communication appears to affect choices and frequencies of reference items when readers are outside the company or the file, direction does not seem to influence these choices and frequencies when intended readers are above, below, and at the same level as the writers. That is not the case, however, in choices and frequencies of grammatical subjects of main clauses. People are the subjects of 70 percent of the T-units written to those above, 62 percent to those below, 55 percent to the same level as the writer and .56 percent to readers outside the office. This preference shows that the writers presume readers share their knowledge of the referents, either from in the text or outside of it. Only 3 percent and 27 percent of the T-units in memoranda addressed to the file and to multiple readers have people as subjects. One would expect this disparity, since, as was noted above, most memoranda directed to the file and to multiple 85 readers deal with inventories and tell readers what they need to known in order to act in the future. Subjects in memoranda written to the file and to multiple readers are mainly nominals and "others" (60 percent and 68 percent, respectively). The memorandum below, addressed to a staff member about to transfer to another office, reflects approximately the proportion and kinds of subjects in memoranda written to staff above the writer. Moreover, Memorandum 503 also reflects the kinds and proportion of verbs used in communication directed above the writer. Memorandum 503: Above the Writer. Subject: Financial Institutions Group Work in the X Area This memo is to familiarize you with our past work and practice development efforts in the X area. [Deleted in Original! S&L' ~ Performed a cash management/float reduction engagement. Total fees were approximately $60,000. A is the client partner. Our primary client contact jLs [Deleted in Original] CFO. [Deleted in Original] S&L Proposed on a cash management/float reduction engagement. Total fee estimate $40-60,000. B _is the client partner. Our primary client contact [Deleted in Original] an ex-ABCer. We have not yet been engaged; estimated proba bility of success on our proposal effort is 25%. Performed a cash management review in connection with the audit. C jis the client partner. Our primary client contact is D, Controller. C, D, E, and I met-with [Deleted in Original] Treasurer, to discuss our management consulting capabilities. E, X Audit Manager, jls the person with ABC that has the best relationship at the client. We have done some pension 86 audit work but they are not a full audit client. X would like to get this audit. MNP S&L Performed a three week profit improvement study; total fees $75,000. We have not been retained for the follow-on implementation work related to this engagement. [Deleted in Original] is in the process of being acquired by EFG. Our plan is to try to impress [Deleted in Original] with our knowledge of problems and the need to retain ABC. B will try to help us in this effort during his work on the acquisition. B has a memo from F detailing our history at [Deleted in Original]. Good luck with your new assignment. If I can be of any help to you, please call. While the use of "you understood" as subject varies little in relation to readers, there is one notable exception: half of the grammatical subjects of T-units addressed to those below the writer are "you understood." Communications addressed to staff above the writer tend to be for-your-information memoranda. As such, the "you understood" subjects are in sentences that ask the reader to acknowledge receipt of the information in the memorandum. As previously mentioned in connection with choices of gram matical subjects, half of the T-units in memoranda to readers below the writers' levels in the firm are commands. This high proportion of commands is understandable when one observes that most memoranda directed to this group of readers do indeed tell readers what they must do. As these excerpts illustrate, combined with recurring should/ought to and must modals, the frequent occurrence of "you understood" reflects the power relationship between the reader and writer. 87 Memorandum 701. (Excerpt) Below the Writer. Examine propriety of bulk storage - It is usually not sufficient to just count items stored in large quantities by reading dials, gauges, or meters. If at all possible [you] physically see that tanks, bins, and silos contain approximately the amount the gauge indicates and that the storage contains what the client represents is in the storage. Sometimes formulas calculating weight or volume may be required to determine quantity. Chemical analysis may be necessary to determine the contents of storage so it may be necessary to take a sample. If the bulk storage container came directly from a vendor and is sealed (such as a railcar) [you] examine the unbroken seal and [you] make note of identification number for future audit testwork. Complete required documentation - Additional information as required on the inventory observation questionnaire should be completed for every engagement. An inventory observation memo should also be prepared for every engage ment. Other items which need to be documented in addition to testwork performed (such as testcounts) may include ASB 1980-6 sample size calculations and SEADOC safeguard control worksheet. [You] Make sure you know how many testcounts are needed in working papers and whether your test counts should emphasize certain items over others. Elements of Coherence Direction of communication also influences how writers show relationships between propositions. The frequencies and distribu tion of conjunctions and sentence openers are shown in Table 12. Memoranda written to readers above the writers have the fewest conjunctions. In part, the brevity of these memoranda accounts for the few conjunctions. (The average number of T-units in the 9 memoranda in this group is 6.3, and the lengths range 1 T-unit to 21. See Table 25 on page 119 for comparisons with other groups.) In addition, some connections are made through adverbial sentence openers. More interesting, however, is the writers' use of 88 formatting to relate ideas. As Memorandum 505 illustrates, writers list, indent, double space, and provide headings to show relationships. Memorandum 505. Above the Writer. Subject: [Deleted in Original] Positive aspects of the [Deleted in Original] • Development of professional relationships with audit personnel • Appreciation of auditing procedures, problems, and concerns • Working knowledge of workpaper techniques • Understanding of how consulting opportunities can be identified during an audit. [Deleted in Original] However, in the future, some changes could be made to strengthen the program. If made, I feel that these changes could further enhance the program and reduce some misunderstanding that developed during my period as a [Delete in Original] • A memo introducing the [Deleted in Original] as well as the candidate to both the audit and consulting departments. This should promote a sense of belonging and avoid confusion among audit and consulting professional staff with respect to the program This is necessary due to the newness of the program; as the program becomes more widely understood, only the candidate need be intro duced . • [Deleted in Original] should be scheduled in audit for the first and third three-month period in the program and consulting the second and fourth. This should promote: continuity in the audit learning curve 89 a sense of belonging the development of professional relationships more rapidly. • Audit management group personnel must be made aware that [Deleted in Original] have long-term goals residing in X. The above mentioned structural changes should help to strengthen the [Deleted in Original]. Even though the very nature of the program does not easily lend itself to structure, I feel that if we can develop some loose framework, everyone involved will benefit. Memoranda to the file, unlike those written to readers above the writer, are long, ranging from 18 to 80 T-units and averaging 44 T-units. They, too, have few conjunctions and sentence openers to show relationships, partially because the writers use headings, lists, and spacing. In addition, in most accounting firms, working papers assume established structures, of at least an overview, general controls, and specific controls; readers expect those relationships to exist in working papers. 90 TABLE 12 DIRECTION OF COMMUNICATION AND ELEMENTS OF COHERENCE: FREQUENCIES OF ELEMENTS OF COHERENCE IN MEMORANDA IN T-UNITS (T-U) (T-Units Per Category Shown in Parentheses) Reader Above (57 T-U) Reader Below (60 T-U) Reader at Same Level (153 T-U) To File (220 T-U!) Reader Outside (65 T-U) Multiple Readers (240 T-U Con iunctions Additives .02 . 17 .09 .03 .11 .03 Adversatives .05 .00 .03 ■ .02 .02 .01 Causals .00 .03 .02 .02 .05 .06 Temporals .0*0 .00 .01 .02 .00 .02 Total .07 .20 .15 .09 .18 . 12 Sentence Openers Adverbials .23 .33 .23 . 11 .28 .22 Verbals .00 .03 .03 .02 .02 .02 Inverted Constructions .02 .02 .07 .00 .03 .02 Total .25 .38 .33 . 13 .33 .26 From the figures in Table 12, one might infer that elements of coherence occur most frequently in memoranda written to readers below writers. However, one long (46 T-units) memorandum accounts for the unusually high number of conjunctions and sentence openers. It is, according to the writer, "a discussion of major procedures and why they are performed." Therefore, it was necessary for the writer to express cause-and-effeet relationships (see Memorandum 701) . In general, when the frequencies of sentence openers in the memoranda are examined in relation to intended readers, they con- 91 trast with those that Christensen found in his writers' discourse. He observed that about 25 percent of his writers used openers. Accountants, however, used openers in only 11 percent of the memo randa directed to the file and in 38 percent of those written to readers below the writer. Features That Imply Reader-Writer Relationships As the figures in Table 13 illustrate, the number of hedges in the writing sample is low (.12 per T-unit). /TABLE 13 DIRECTION OF COMMUNICATION AND READER-WRITER RELATIONSHIPS: FREQUENCIES OF READER-WRITER RELATIONSHIP ITEMS IN MEMORANDA IN T-UNITS (T-U) (T-Units Per Category Shown in Parentheses) Reader Above (57 T-U) Reader Below (60 T-U) Reader at Same Level (153 T-U) To File (220 T-U) Reader Outside (65 T-U) Multiple Readers (240 T-U) Hedges Approximator/Adaptors .04 .00 .02 .05 .03 .02 Approximator/Rounders .04 .02 .08 .02 .03 .05 Plausibility/Shields .08 .03 .05 .03 .02 .02 Attribution/ShieIds .04 .00 .01 .00 .02 .03 Total .20 .05 .16 .10 .10 .12 Modals Can/Could .02 .02 .02 .01 .00 .01 May/Might .00 .05 .01 .00 .03 .00 Must .02 .03. .00 .00 .02 .00 Other .02 .00 .02 .00 .00 .05 Negative .07 .05 .06 .07 .09 .08 Should/Ought To .07 .12 .10 .04 .11 .06 Will/Would .09 .08 .14 .07 .09 .20 Total .29 .35 .35 .19 .34 ‘ .40 Personal Pronouns First-person Singular .30 .10 .16 .05 .22 .15 First-person Plural .35 .03 .27 .05 .35 .14 Second-person Pronouns .35 .78 .63 .20 .48 .10 Third-person Pronouns . 17 .13 .10 .03 .19 .21 Total 1.17 1.04 1.16 .33 1.24 .60 Formulary Expressions Future Contact .03 .00 .04 .00 .05 .01 Request Acknowledgment .03 .00 .05 .00 .02 .00 Solicitous C o m m e n t .02 .03 .00 .00 .02 .00 Total .08 .03 .09 .00 .09 .01 93 While direction of communication does not greatly influence uses of hedges, as one would expect, memoranda written to readers above and at the same levels as the writers contain the most hedges (.20 and .16 per T-units). Nor is it a surprise that memoranda written to readers below the writer contain the fewest hedges, since the group of five memoranda consists of four recommendations and one description of an inventory procedure readers are to follow. In general, direction of communication does not seem to affect the number or kinds of modals writers use, although memoranda directed to the file contain only .2 modals per T-unit in contrast to the overall use of .3 per T-unit because working papers and summaries of clients' histories require few modals. Among the stylistic features analyzed in this study, direction of communication affects most dramatically the writers' choices of pronouns. Memoranda written to readers above the writer reflect a 2 to 1 preference for first-person pronouns over second person. Since writers of those memoranda usually aim to describe past events or accomplishments or transmit.information to a superior, the nature of the contents seems to determine the choices of pronouns. The most surprising relationship between first- and third- person pronouns appears in memoranda to readers outside the office. Traditionally, writers in business have been admonished to write with the reader in mind. Specifically, they are told to use "you" rather than "we" to create reader-oriented correspondence. In communications to outside readers, accountants used 6 first-person 94 to 5 second-person pronouns, a preference that can in part be accounted for by the fact that all but 4 of the memoranda in this category are addressed to readers in the firm but out of the office of the writers. Thus, there is no need to maintain the goodwill of the reader, which is the rationale for "you" versus "we" orientation of memoranda directed to outside readers: Memorandum 406. Outside the Office. Subject: [Deleted in Original] As in prior years, we anticipate that we will require the assistance of your office in observing the physical inventories for [Deleted in Original] operations. How ever, in a departure from past practice, the client has tentatively scheduled the inventory for February 24 through February 26, 1982, instead of the end of January, its fiscal year-end. Accordingly, we are notifying you at this time to facilitate scheduling; further instructions regarding locations, timing, and client contacts will be forwarded as soon as client arrangements are finalized. If you have any questions, [you] please contact me at (213) 553-1280. To acknowledge receipt of this memo, [you] please sign and [you] return the copy attached. Direction of communication also affects pronoun choices in memoranda written to multiple readers. They contain more first-person pronouns (.29 per T-unit) than second-person pronouns (.10 per T-unit). That ratio is partially attributable to the fact that the memoranda to multiple readers have as one of the readers a member of the firm above the writer. Those written to readers below the writer contain 1 first-person pronoun to 6 second-person 95 pronouns, a ratio that can in part be attributed to the many "you understood" sentences in the memoranda in this category. There appears to be no unusual pattern in the uses of formulary expressions based on who the readers are. As one would expect, memoranda written to the file contain none since those documents neither require a response from the reader nor attempt to maintain the goodwill of the reader. Moreover, one would expect that a writer addressing a manager above him would suggest he was available to discuss the contents of the memorandum, as the writer does, in this sentence: If you have any questions, please feel free to call on me. Additionally, one would expect that a writer addressing a manager above him would couch requests for acknowledgments of materials in formulary expressions, expressing the request as routine, as in Please sign and return the attached copy of this memoran dum. The single interesting observation about formulary expressions is their rarity in memoranda addressed to readers below or at the same level as the writers of those memoranda. Only 2 out of 19 memoranda close with formulary expressions. Conclusions Direction of communication appears to influence writers' choices of the stylistic features analyzed in this study. In 96 memoranda directed to readers at the same level and above the writer, choices and frequencies of those features are generally the same. The single difference is that in memoranda written to readers above the writer more first-person pronouns are used than second-person. The relationship is reversed in memoranda directed to readers at the same level. Memoranda to readers below the writer differ stylistically from those directed to other levels; they contain fewer hedges and more command sentences. Memoranda to the file have predominantly nominal and "other" grammatical subjects, few sentence openers, and no formulary expres- s ions. Memoranda directed to readers outside the office of the writer, whether to readers in the same firm as the writer or to readers outside the company, have more reference items and more formulary expressions than other memoranda have. Taken as a whole, the memoranda directed to multiple readers have few formulary expressions, the fewest sentences with people as subjects, and the fewest command sentences. Individually, however, style seems to be influenced by the specific readers that the writers identified. 97 CHAPTER VII SUMMARY OF FINDINGS, CONCLUSIONS, AND IMPLICATIONS FOR TEACHING AND RESEARCH Scholars such as Milic, Hiatt, and Garnlca claim that we should be able to describe an individual’s writing style in terms of his or her linguistic choices. Furthermore, composition researchers like Odell, Goswami, Williams,• and Hake have shown that linguistic choices are determined by the constraints of professional environment, subject matter, purpose, and the writer's perceptions of the reader. In general, this study sought to determine if the claims of literary stylists are valid for nonacademic writers and in nonacademic writing and if the conclusions of composition researchers are valid for writers of accounting memoranda. Specifically, the study sought to determine if linguistic choices made by accountants are affected by their decision style preferences, the purposes of discourse, and the directions of communication. The definition of writing style used in this study was formulated from several assumptions that underlie Louis Milic's theory of style: A writer's style is conditioned by the context in which he writes, the linguistic choices available in his language, the effect he desires, and some unconscious predisposition toward 98 certain linguistic features. To these assumptions about style I added the assumption that a writer's intended audience affects style. These extratextual influences on writers' linguistic choices provide the organization for the remainder of this chapter. The Context for Written Discourse In Addressing An Audience: A Study of Expert-Novice Differences in Writing, Marshall A. Atlas urges composition researchers to standardize the knowledge which subjects bring to research tasks. Research designs similar to those developed for the doctoral studies discussed in Chapter II reflect a growing awareness that—writing processes and— products must be Controlled" if the research is to contribute to our understanding of writing processes and strategies. The memoranda investigated in this study represent a writing sample in which the context, or scene, of the writing was rigorously controlled. In the lexicon of Kenneth Burke's dramatistic theory of language,108 the scene-act ratio, in which modes of response (acts) are viewed as motivated by the situation in which they take place (scene), was fixed. In Lloyd Bitzer's terms, the seven writers shared rhetorical situations: All junior-level accountants in a Big Eight accounting firm, the writers were constrained by the "people, events, objects, and relations which are parts of the situation because they have the power to constrain the decisions and actions needed to modify the exigence"109 that prompt the writers to compose the memoranda. And in sociolinguistic terms, the writers 99 shared contextual knowledge. As Marilyn Cooper explains in "Context as Vehicle: Implicatures in Writing," The only chance writers have of communicating their meanings is by assuming that potential readers have largely the same knowledge of those facts about the world relevant to their meanings as they themselves do, and that readers will, when faced with a text (an object that is always multiply ambiguous), use the same interpretive strategies that the writers would.110 Context is important to writers on the job, as Dixie Goswami and Lee Odell have reported in several sources. In their study of writing in three settings, they found that writers draw from their professional environments for strategies in composing texts. In this study of writing in a fourth setting, a Big Eight accounting firm, the conclusions are consistent with Goswami and Odell's and support their recommendation that composition students be given full rhetorical contexts for their classroom writing tasks so that they will be better prepared to write on their jobs. Because the context in which the memoranda were written was so deliberately controlled, one must be cautious in generalizing about the effects of decision style, purpose, and direction of communica tion on the linguistic choices reflected in writing from other concexts. However, the findings of the study raise several questions about the effect of context on writers' linguistic choices, questions that should be addressed in future research. To what extent does shared knowledge affect the writers' linguistic choices? 100 How do the linguistic choices made by the junior-level accountants compare to those made by, for example, upper- level management? How do the linguistic choices made by the junior-level accountants compare to those made by writers in other professions? Linguistic Choices Linguistic behavior does not consist simply of producing separate sentences; instead, it consists of using sentences to create discourse. Therefore, writers are obligated to select linguistic devices to create cohesion if they hope to communicate effectively. They are further obligated to select linguistic -elements -to -help readers -perceive intended relationships "between adjacent propositions; that is, they must create coherent discourse. Additionally, because discourse is created for an audience, writers must choose words and phrases that imply relationships between themselves and their intended readers. Since no apparatus existed for comparing linguistic preferences among decision stylists, purposes of discourse, and directions of communication, it was necessary to develop a discourse analysis framework. Influenced largely by M. A. K. Halliday's "Notes on Transitivity and Theme in English, Parts 1, 2, and 3," Halliday and Hasan's Cohesion in English, the work of Francis Christensen and Joseph Williams, and insights from sociolinguists like George Lakoff and Robin Lakoff, the framework used in this study isolates the following linguistic features for analysis: reference items, grammatical subjects, verbs and verb voices, conjunctions, sentence 101 openers, hedges, modals, pronouns of address, and formulary expressions. Cohesion Elements For this study, the selected cohesive elements of reference items, grammatical subjects, and kinds of verbs and voices were one category of linguistic features used to identify writing style. As can be seen from the figures in Tables 14, 15, and 16, choices of these devices and the frequencies with which they appear in the memoranda are influenced to varying degrees by writers' decision styles, purposes of discourse, and directions of communication. .Although-specif ic -relationships between~those variables and-cohesive devices are discussed in Chapters IV, V, and VI, respectively, some general relationships are displayed in Tables 14, 15, and 16. Generally, reference items seem to be influenced by the three variables, although the more dramatic differences can be seen within stylistic choices of items, as Table 14 illustrates. 102 TABLE 14 COHESIVE ELEMENTS: REFERENCE ITEMS PER T-UNITS Personal Pronouns Possessive Determiners The Demonstratives Totals Decision Stvle Analytic-Directive 1. 13 .23 .98 . 14 2.48 Analytic-Conceptual .25 . 16 .95 .13 1.49 Conceptual-Directive .35 .31 .96 .13 1.75 Conceptual-Analytic .20 . 15 .71 . 11 1.17 Purpose of Discourse Description of Procedure Z . 19 . 16 .91 .09 1.35 Recommendation .51 .29 1.01 . 18 1.99 Request 1.00 .30 .88 .06 2.24 Document a t ioii .39 — .81 .15 1.54 Description/ Documentat ion .28 . 19 .88 . 10 1.45 Direction of Communication Reader Above .61 .35 .73 . 16 1.85 Reader Below .53 .08 .83 . 12 1.56 Reader at Same Level .59 .31 1.00 . 18 2.08 To the File .07 .10 .94 .09 1.20 Outside the Office .60 .37 1. 10 . 15 2.22 Multiple Readers .34 .21 .86 . 12 1.53 For instance, in the case of purpose, discussed in Chapter V, requests contain twice as many personal pronouns and one-third as many demonstratives as recommendations do. And in the case of direction of communication where memoranda to the file contain the fewest and memoranda to readers outside the office contain the most, 103 there are again substantial differences among the preferences for reference items (see Chapter VI for discussions of these phenomena). In addition, as the figures in Table 15 suggest, choices of grammatical subjects differ among decision stylists, purposes, and directions of communication. In the case of decision styles, differences are small when totals are considered. Nevertheless, Conceptual-Analytic stylists, for example, use the fewest "you understood's," pronouns, and names of people as grammatical subjects, and the most nominals; "people" subjects appear most frequently in requests and in memoranda written to multiple -readers. Moreover, "you understood" is used as the grammatical subject ten times more frequently in memoranda to readers below the writer than in memoranda to multiple readers. 104 TABLE 15 COHESIVE ELEMENTS: GRAMMATICAL SUBJECTS PER T-UNITS You Jnderstood Pronouns (Excluding Personal) Names and Personal Pronouns Nominals Others Total Decision Style Analytic-Directive . 18 .06 .44 .22 . 17 1.07 Analytic-Conceptual • .26 .05 . 15 .29 .22 .97 Conceptual-Directive .16 .03 .32 .25 .28 1.04 Conceptua1-Analyt ic .05 .02 . 12 .53 .27 .98 Purpose of Discourse ■ Description of Procedure .23 .04 ' .14 .33 .24 .98 Recommendat ion .25 .06 .25 .30 .18 1.04 Request _____ _____ __.22___ — - -.03 - -- ----: 42'-- -- .18 — . 15' T00"‘ Documentation .01 .02 .29 .33 .32 .97 Description/ Documentation .03 .05 .27 .41 .26 1.02 Direction of Communication Reader Above . 18 .05 .53 .18 . 18 1.12 Reader Below .50 . 10 . 12 . 17 . 12 1.01 Reader at Same Level .28 .03 .27 .27 .21 1.05 To the File .20 .04 . 11 .33 .27 .95 Outside the Office .22 .02 .34 .29 . 18 1.05 Multiple Readers .05 .03 .22 .40 .28 .98 The figures in Table 16 also show some large differences in choices and frequencies of verbs and voices based on writers’ decision styles, purposes of discourse, and directions of communication. 105 TABLE 16 COHESIVE ELEMENTS: KINDS OF- VERBS AND VOICES PER T-UNITS Passive Voice Linking Active Intransitive You Understood Total Decision Style Analytic-Directive .21 .19 .42 .07 . 18 1.07 Analytic-Conceptual .27 .16 .26 .04 .26 .99 Conceptual-Directive .27 .32 .40 .06 .16 1.21 Conceptual-Analytic .39 . 14 .37 .02 .05 .97 Purpose of Discourse Description of Procedure .28 . 11 .35 .04 .23 1.01 Recommendation .3° . 18 .28 .03 .25 1.04 Request .24 . IS .39 .06 .21 1.05 Documentat ion .28 .24 .33 .06 .07 .98 Description/ Documentation .26 .26 .38 .07 .03 1.00 Direction of Communication Reader Above . 18 .32 .29 .02 .18 .99 Reader Below . 17 .12 . 17 .03 .50 .99 Reader at Same Level .30 .17 .27 .03 .28 1.05 To the File .27 . IS .35 .05 .20 1.02 Outside the Office .29 .05 .45 .02 .22 1.03 Multiple Readers .34 . 18 .38 .06 .05 1.01 Conceptual-Analytic stylists, for instance, use one-fifth as many commands as Analytic-Conceptual stylists use. Documentations and description/documentations contain more linking verbs than other discourse purposes contain and fewer commands. Furthermore, linking verbs are used most frequently in memoranda written to readers above the writers and least frequently to readers outside the office. 106 Passive voice verbs, used frequently in institutional prose to avoid "i" or "we," (and, thus, to avoid responsibility) appear seldom in memoranda written to readers above the writers. Since passive voice verbs require more time to process than active voice verbs do, perhaps the value of the readers' time is a greater constraint on the writer than is the manager-subordinate relationship. From this speculation arises a question for future study. Is there a hierarchy among constraints on the writer? To ascertain whether or not such a hierarchy exists is, of course, beyond the scope of this study. A larger corpus than the 67 memoranda analyzed here and a framework for describing context would be necessary. Nevertheless, to be able to explain phenomena inconsistent with expected patterns could be helpful. The importance of cohesive devices to clue readers in the relationships of propositions throughout stretches of discourse is well established in our composition pedagogy. Generally, however, when exercises to promote the use of various cohesive devices are offered, no attention is paid to the kinds of discourse that serve as examples. That we find distinctions among preferences for cohesive devices in a writing sample as controlled as this demoristrates that we need to be more cautious in generalizing about uses of cohesive ties. For example, in the memoranda of accountants, documentations and description/documentation contain twice as many linking verbs as the other purposes contain. These 107 distinctions suggest that our writing instruction should focus on purposes of discourse and that cohesive ties should be analyzed in terms of their recurrence in discourse intended for specified purposes. Although we must isolate additional cohesive devices (such as lexical repetition) in larger writing samples before we can say unequivocally that purpose of discourse and the direction of the communication help to determine writers' choices of cohesive items, the findings of this study support that contention. Moreover, it is important to look at cohesive features in requests, descriptions of procedures, etc., from writers in other business contexts to con trast their preferences for cohesive devices to the preferences of the accountants who participated in this study. Elements of Coherence As would be expected, choices and frequencies of elements of coherence analyzed in this study were generally less influenced by the writers' decision styles, purposes of discourse, and directions of communication than were choices and frequencies of cohesive devices, as the figures in Tables 17 and 18 illustrate. 108 TABLE 17 ELEMENTS OF COHERENCE: CONJUNCTIONS PER T-UNITS Additives Adversatives Causals Temporals Total Decision Style Analytic-Directive . 11 .04 .04 .01 .21 Analytic-Conceptual .06 .01 .03 .01 .14 Conceptual-Directive .05 .04 .01 .03 .13 Conceptual-Analytic' .02 .00 .05 .00 .07 Purpose of Discourse Description of Procedure .06 .02 .04 .01 .13 Recommendation .07 .02 .05 .01 .15 Request .09 .03 .00 .00 . 12 Documentat ion .03 .02 .01 .02 .08 Description/ Documentat ion .03 .02 • .02 .00 .07 Direction of Communication Reader Above .02 .05 .00 .00 .07 Reader Below . 17 .00 .03 .00 .23 Reader at Same Level .09 .03 .02 .01 . 15 To the File .03 .02 .02 .02 .09 Outside the Office . 11 .02 .05 .00 .18 Multiple Readers .03 .01 .06 .02 . 12 109 TABLE 18 ELEMENTS OF COHERENCE: SENTENCE OPENERS PER T-UNITS Adverbials Verbals Inverted Constructions Total Decision Stvle Analytic-Directive .30 .03 .08 .41 Analytic-Conceptual » .20 .01 .02 .23 Conceptual-Directive .20. .03 .01 .24 Conceptual-Analytic . 11 .04 .03 .18 Purpose of Discourse Description of Procedure . IB .02 .01 .21 Recommendation .27 .02 .03 .32 Request .24 .00 .09 .33 Documentation . 14 .02 .02 . 18 Description/ Documentat ion . 16 .03 .01 .20 Direction of Communication Reader Above .23 .00 .02 .25 Reader Below .33 .03 .02 .38 Reader at Same Leve 1 .23 .03 .07 .33 To the File . 11 .02 .00 . 13 Outside the Office .28 .02 .03 .33 Multiple Readers .22 .02 .02 .26 110 Some differences in the uses of conjunctions and sentence openers are notable, however. For instance, as discussed in Chapter IV, memoranda written by Conceptual-Analytic stylists contain one-third fewer conjunctions and sentence openers than memoranda written by Analytic-Directive stylists. Furthermore, the fewest conjunctions appear in documentations and description/ documentations and in memoranda to the file and to readers above the writers. Sentence openers appear in equal numbers in descriptions of procedures, documentations, and description/documentations and in memoranda written to the file. Our traditional reliance on lists of transitions to teach students how to create coherent discourse has been rejected by many composition teachers. The findings related to preferences for elements of coherence support their assessment. Perhaps coherence should instead be taught in terms of audience for the discourse since that component of the rhetorical situation influenced accountants' choices of sentence openers and conjunctions. In addition, although findings in the small sample of writing examined here are inconclusive, perhaps more attention should be given to the role of formatting (headings, space, indentations, etc.) since accountants frequently depend on those devices to achieve coherence. Coherence should continue to be a concern for composition researchers. For instance, as was mentioned previously, sociolinguists have told us that knowledge shared by the speaker/writer and hearer/writer contribute to making discourse Ill coherent. While on several occasions I speculate that shared knowledge may be a contributing factor to the presence (or absence) of elements of coherence, no methodical examination was made in this study to determine the extent to which shared knowledge contributes to the frequency of conjunctions and sentence openers in the memoranda. The findings of this study indicate that such analyses are needed. The memoranda analyzed in this study represent a writing sample from a group of writers who share rhetorical situations. The accountants can even assume that readers in their firm share their education backgrounds and professional experiences. One would assume, therefore, that memoranda written to readers within the firm would contain equal numbers of conjunctions and sentence openers. Judging from the number of coherence elements used, that is not the case, however; readers below the writer, for instance, seem expected to be less knowledgeable than their co-workers at other levels in the firm. Conversely, one would assume that the few memoranda written to readers outside the firm would reflect less shared knowledge, that is, would contain more conjunctions and sentence openers. That is not the case either. These findings are pertinent to the issue of rater reliability: When raters judge a writer's success in achieving his or her intended purpose--whether the judgment is being made for research purposes or for pedagogical reasons--the raters' familiarity with the writer's context, audience, subject matter, etc., will surely 112 influence their assessment (see discussion below, in Purpose and Writing Style). Items That Imply Reader-Writer Relationships As the figures in Tables 19, 20, 21, and 22 suggest, choices and frequencies of items that imply reader-writer relationships may be influenced by writers' decision style preferences, purposes of discourse, and directions of communication. Among decision stylists, for instance, Conceptual-Analytic stylists prefer to use hedges (Table 19) and modals (Table 20) more frequently than personal pronouns and formulary expressions to imply relationships with their readers. Analytic-Directive stylists, on the other hand, prefer to use formulary expressions (Table 22) and personal pronouns (Table 21) instead of hedges and modals (see Chapter IV for discussion). 113 TABLE 19 READER-WRITER RELATIONSHIP ITEMS: HEDGES PER T-UNITS Approximator/ Adaptors Approximator/ Rounders Plausibility Shields Attribution Shields Total Decision Style Analytic-Directive .01 .07 .09 .01 .18 Analytic-Conceptual j.02 .03 .02 .01 .08 Conceptual-Directive .05 .06 .04 .02 .17 Conceptual-Analytic .02 .05 .02 .03 .22 Purpose of Discourse Description of Procedure .02 .02 .01 .01 .06 Recommendation • 02 .06 .03 .01 .12 Request .03 . .03 . 18 .06 .30 Documentation .05 .06 .05 .02 . 18 Description/ Documentation .05 .04 .02 .01 . 12 Direction of Communication Reader Above .04 .04 .08 .04 .20 Reader Belou .00 .02 .03 .00 .05 Reader at Same Level .02 .08 .05 .01 .16 To the File .05 .02 .03 .00 .10 Outside the Office .03 .03 .02 .02 . 10 Multiple Readers .02 .05 .02 .03 . 12 114 TABLE 20 READER-WRITER RELATIONSHIP ITEMS: MODALS PER T-UNITS / Can/ Could May/ Might Must Other Negative Should Ought To Will Would Total Decision Style Analytic-Directive .01 .00 .01 .01 .07 .04 .19 .33 Analytic-Conceptual .01 .01 .01 .03 .06 .08 .11 .31 Conceptual-Directive .02 .02 .01 .01 .08 .08 .05 .27 Conceptual-Analytic .01 .01 .00 .01 . 10 .05 .21 .38 Purpose of Discourse • Description of Procedure .01 .02 .00 .01 .07 .03 . 15 .29 Recommendation .01 .00 .02 .05 .04 . 19 . 12 .43 Request .00 .00 .00 .00 .06 .00 .02 .18 Documentation .02 .01 .01 .02 .09 .03 .08 .26 Description/ Documentation .01 .01 .00 .01 .08 .02 . 13 .26 Direction of Communication Reader Above .02 .00 .02 .02 .07 .07 .09 .29 Reader Below .02 .05 .03 .00 .05 . 12 .08 .35 Reader at Same Level .02 .01 .00 .02 .06 . 10 . 14 .35 To the File .01 .00 .00 .00 .07 .04 .07 . 19 Outside the Office .00 .03 .02 .00 .09 . 11 .09 .34 Multiple Readers .01 .00 .00 .05 .08 .06 .20 .40 115 TABLE 21 READER-WRITER RELATIONSHIP ITEMS: PERSONAL PRONOUNS PER T-UNITS First Singular First Plural Second Singular Plural Third Singular Plural Total Decision Style Analytic-Directive .46 .27 .63 .29 1.65 Analytic-Conceptual . 10 .09 .37 .06 .62 Conceptual-Directive .10 .30 .34 .09 .83 Conceptual-Analyt ic .03 20 . 10 .12 .45 Puroose of Discourse Description of Procedure .05 . 15 .28 .08 .56 Recommendation .22 . 16 .53 .15 1.06 Request .52 . 18 .48 .30 1.48 Documentat ion . 12 . 16 . 15 . 18 .61 Description/ Documentat ion .09 . 18 .06 . 15 .48 Direction of Communication Reader Above .30 .35 .35 . 17 1-.17 Reader Below . 10 .03 .78 .13 1.04 Reader at Same Level . 16 .27 .63 . 10 1. 16 To the File .05 .05 .20 .03 .33 Outside the Office .22 .35 .48 . 19 1.24 Multiple Readers . 15 . 14 . 10 .21 .60 116 TABLE 22 READER-WRITER RELATIONSHIP ITEMS: FORMULARY EXPRESSIONS PER T-UNITS Future Contact Acknowledgment Request Solicitous Comments Total Decision Style Analytic-Directive .04 .02 .01 .07 Analytic-Conceptual .01 .01 .01 .03 Conceptual-Directive .02 .01 .01 .04 Conceptual-Analytic .00 .02 .00 .02 Purpose of Discourse Description of Procedure .01 .00 .00 .01 Recommendat ion .03 .02 .00 .05 Request .03 .03 .00 .05 Document at ion .01 .02 .00 .03 Description/ Documentation .01 .00 .00 .01 Direction of Communication Reader Above .03 .03 .02 .08 Reader Below .00 . 00 .03 .03 Reader at Same Level .04 .05 .00 .09 To the File .00 .00 .00 .00 Outside the Office .05 .02 .02 .09 Multiple Readers .01 .00 .00 .08 117 Additionally, it is not surprising that items that imply reader-writer relationships appear most frequently in requests and recommendations. Moreover, with the exception of modals, over half of the reader-writer relationship items in the 67 memoranda appear in memoranda that request or recommend. And, in general, items that imply reader-writer relationships appear least frequently in memoranda written to the file. Specifically, hedges (Table 19) are used four times more frequently in memoranda written to readers above the writers than to readers below. Discourse Length The findings of this study challenge the common textbook prescription that conciseness is preferable to lengthiness. Writers of business correspondence have traditionally been admonished to be concise, to save the reader's time. Recently, "conciseness" has come to be measured in tqrms of readers' needs and demands of purpose and subject matter, in terms of what E. D. Hirsch calls "relative readability." Discourse length, then, is in part the result of writers' choices. Thus, to some degree, length is a matter of style. And it may also be a result of the writer's adhering to the Gricean maxim of Quantity; that is, the writer should tell the reader what he or she needs to know and no more (see note 30). In the memoranda of accountants, as Table 23 illustrates, lengths of memoranda do not appear to be influenced by writers' decision style preferences. For instance, while both of the two 118 Conceptual-Directive stylists (4 and 5) wrote 7 memoranda, one wrote 18.1 T-units per memorandum and the other wrote 9.9 T-units per memorandum. Furthermore, the lengths of their memoranda ranged from 5 to 40 T-units and from 3 to 21, respectively, suggesting that the length of a memorandum is probably not determined by writers’ decision styles. TABLE 23 DISCOURSE LENGTH AND DECISION STYLE Decision Styles Accountant Number of Memoranda Total T- Units Average T-Units Range Analytic-Directive 1 1 23 23 0-23 Analytic-Directive 2 16 67 4.2 1-10 Analytic-Conceptual 3 12 76 6.3 3-26 Analytic-Conceptual 7 13 300 23.1 2-57 Conceptual-Directive 4 7 127 18.1 5-40 Conceptual-Directive 5 7 67 9.9 3-21 Conceptual-Analytic 6 11 133 16.2 1-32 The purpose of discourse and the direction of communication, however, do seem to influence the lengths of memoranda. As one can infer from Table 24, memoranda that describe procedures are on the average four times longer than requests that require responses. They range from 3 to 80 T-units, whereas requests range from 3 to 8 T-units. 119 TABLE 24 DISCOURSE LENGTH AND PURPOSE OF DISCOURSE Purpose Number of Memoranda Total T- Units Average T-Units Per Memoranda Range Description of Procedure 15 349 23.3 3-80 Recommendat ion 28 204 7.3 1-40 Request 6 33 5.5 3-8 Docueentation 15 195 13.0 1-57 Description/ Document at ion 16 301 18.8 1-57 As the figures in Table 25 indicate, memoranda written to readers above the writer are the shortest, ranging from 1 T-unit to 21 T-units. The widest range of lengths is in those written to the file (18 to 80 T-units). TABLE- 25 DISCOURSE LENGTH AND DIRECTION OF COMMUNICATION Intended Readers Number of Memoranda Total T- Units Average T-Units Per Memoranda Range Above 9 57 6.3 1-21 Below 5 60 12.0 1-46 Same Level 21 153 7.3 3-40 To the File 5 220- 44.0 18-80 Outside Office 8 65 8.1 4-13 Multiple Readers 19 240 12.6 1-36 This study suggests we should consider the length of a piece of discourse to be dependent at least on the purpose and audience for whom it is intended. In short, we can no longer feel confident that 120 telling students to be concise is always in the best interest of effective communication. Because this study was limited to a small number of accountants and was designed to describe the 67 memoranda in this corpus, the generalizations about linguistic preferences are, of course, limited to the accountants' memoranda. Furthermore, as I have noted fre quently, the findings, conclusions, and implications for teaching and further research discussed here are circumscribed by my choices of linguistic features to analyze. Nevertheless, the discourse analysis framework can be applied to all kinds of written discourse. Since the framework has served to demonstrate that writers' decision styles, purposes of discourse, and directions of communication substantially influenced the linguistic choices of the accountants, researchers can use the framework to contrast the linguistic pre ferences of the junior-level accountants with those of other on- the-job writers. By analyzing products written in many professional environments and fulfilling different tasks, composition teachers may eventually be able to say with assurance that specific purposes of discourse and specific readers require specific linguistic features to be effective. Purpose and Writing Style In the procedure used to collect the writing sample (see Chapter III), accountants were asked to identify the purpose of each of their memoranda. For the stylistic analyses of the memoranda, purposes were identified by trained raters. The results of these 121 two processes, displayed in Table 26, raise questions about composition teaching and research. TABLE 26 PURPOSES OF MEMORANDA IDENTIFIED BV THREE TRAINED READERS AND THE WRITERS OF THE MEMORANDA Writer's Readers' Writer s Readers' Memoranda Intention Purpose Memoranda Intention Purpose 101 1,4 4 4 i 407 1 4 4 l 201 2 2 2 3 501 1 4 4 4 202 2 2 2 2 502 4 3 3 2 203 2 2 2 2 503 4 4 4 1 204 2,1 2 2 2 504 3 1 1 4 205 4,2,1 2 2 4 505 2 2 2 1 206 4,2 3 3 3 506 4 1 1 4 207 4 4 4 3 507 1 1 1 1 208 2 2 2 1 601 4 1 1 4 209 4 1 1 4 602 4 4 4 1 ! 210 2,4 2 2 2 603 2 2 2 1 211 4 2 2 3 604 2 4 4 3 212 4,1,2 4 4 4 605 4 3 3 4 213 4 4 4 1 606 2 2 2 1 214 1 4 4 1 607 4 3 3 1 215 1,3,4 2 2 4 608 2 4 4 4 216 3,2,1 2 2 2 609 4 4 4 1 301 1 2 2 2 610 4 4 4 4 302 2 2 2 1 611 1 1 1 4 303 1 2 2 2 701 1 1 1 1 304 1 2 2 2 702 2 1 1 2 305 2 2 2 1 703 2 2 2 3 306 1 2 2 1 704 4 4 4 1 307 2 2 2 1 705 3 3 3 3 308 2 1 1 2 706 4 1 1 4 309 1 2 2 1 707 3 3 3 3 310 2 2 2 2 708 4 1 1 1 311 1 1 1 1 709 2 4 4 2 312 2 ■ 2 2 1 710 3 2 3 1 401 1 1 1 4 711 0 2 2 1 402 1 1 1 4 712 0 1 1 1 403 3,1 2 2 1 713 2 2 2 2 404 1,4 2 3 4 405 1 2 2 1 406 4 1 2 4 Using a common procedure for rating compositions , the trained readers came to at least a two- thirds agreement on the purposes of all memoranda except three. The writers of those memoranda, however, agreed with the raters' decisions in only 50 percent of the ' 122 purpose identifications. One reason for this disparity may be that the writers felt less constrained to ascribe a single purpose to their memoranda than the trained raters did. In six cases, writers identified memoranda as having two purposes; in four cases, they identified memoranda as having three purposes. In addition, the guidelines adapted for this study were not originally developed to compare purposes assigned by readers to those assigned by the writers. Whatever the reasons for the lack of agreement, it suggests that rating methods that entail shared knowledge may be more appropriate to rating nonacademic discourse. Furthermore, rating methods sensitive to the many varieties of nonacademic discourse should be developed and tested. Because the ten memoranda mentioned above represent nearly 15 percent of the writing sample, these questions arise: Were there situational constraints that led writers to identify multiple purposes? If so, what were they? Knoblauch has already concluded, based on interviews with business clients, that multiple intentions in discourse are common. More rigorous investigations of nonacademic discourse that focus on purpose are needed. Perhaps interviews with the writers, like those Odell and Goswami conducted for their recent study, could help identify those constraints so that as teachers and researchers, we can more confidently deal with purpose in discourse. While the trained readers did not ascribe multiple purposes to any memoranda, as a group they identified 16 memoranda as being both 123 descriptions of procedures and documentation. In all 16 cases, the writers of those memoranda agreed that one of their intentions was to describe or document. Such a pattern suggests that in helping accountants improve their writing skills, we are obligated to include description/documentations in our writing programs. Fur thermore, if this convention exists in accounting, are there con ventions exclusive to the written communication produced in other business contexts? The trained readers did not reach agreement on 3 of the 67 memoranda. The purpose of, Memorandum 710, for example, was identified by the readers to be a description, recommendation, and request. The writer's intention, however, was to request information; he intended the force of the memorandum to be "Can you use our computer audit tools for any of your engagements?" Memorandum 710. No agreement of Purpose. Subject: IBM System 34 Computers at Clients As the attached memo indicates, we now have the capability to use the firm's audit software (S2190) on most client computer files on IBM System 34s. If you feel that any of your engagements could be made more efficient by using computer audit tools (to foot lengthy files, age receivables or payables, recalculate depreciation, select smaller samples statistically, etc.) please contact me. Similarly, the raters identified Memorandum 404 as a recommendation, a request, and a documentation. The writer intended the memorandum to be both a request and description. 124 Memorandum 404. No Agreement of Purpose Dear John: Enclosed are the revised draft financial statements for [Deleted in Original] and the [Deleted in Original] as of December 31, 1981 and for the year then ended. In addition to your approval of the enclosed drafts, we require the following items prior to issuance of the final financial statements: 1. Copies of the February 2, 1982 minutes for the Association Board of Directors meeting and the Joint Meeting of the Executive Committee of the [Deleted in Original] and the [Deleted in Original] 2. Representation letters, [Deleted in Original] 3. Ms. [X]'s related party questionnaire. As you requested, we have sent a draft of the financial statements to Ms. [X] for her approval and additional comments. A copy of our transmittal letter to Ms. [X] is enclosed. We look forward to delivering the final reports in the near future. Best regards. Very truly yours, The raters identified Memorandum 406 as a description, recommendation, and documentation, none of which was the writer's intention. He intended the memorandum to be a request. Memorandum 406. No Agreement of Purpose Subject: [Deleted in Original] As in prior years, we anticipate that we will require the assistance of your office in observing the physical inventories for operations. However, in a departure from past practice, the client has tentatively scheduled the inventory for February 24 through February 26, 1982, 125 instead of the end of January, its fiscal year-end. Accordingly, we are notifying you at this time to facilitate scheduling; further instructions regarding locations, timing, and client contacts will be forwarded as soon as client arrangements are finalized. If you have any questions, please contact me at (213) 553-1280. To acknowledge receipt of this memo, please sign and return the copy attached. Memoranda whose purposes cannot be agreed upon by trained writers can provide insights into writers' communicative problems. They can, for instance, be isolated from other memoranda in writing samples and used to diagnose writers' needs. In addition, memoranda whose purposes are not clear to trained readers (like those reproduced above) can be used to show w'riters which of their linguistic choices have obscured their purpose. Direction of Communication and Writing Style Nearly a third of the memoranda in the sample are addressed to readers at the same level as the writers. Twenty-one are directed exclusively to peers within the firm, and 14 of the 19 memoranda addressed to multiple readers are addressed to peers. Nevertheless, junior-level accountants write to a variety of audiences, so courses that prepare accountants to write on the job must specify a variety of readers in writing tasks. While accountants write to a variety of readers, there seems to be a pattern in the kinds of memoranda directed to the various levels. For instance, readers below and at the same level as the 126 writers receive more recommendations than any other kind of purpose, while readers above the writer receive more documentations. Only descriptions and documentations are directed to the file, and no requests go outside the office. All readers receive descriptions of procedures, but only readers above and at the same level as the writer receive requests. This finding has implications for composition instruction. For example, the effectiveness of writing programs for accountants can be increased if readers are specified for writing tasks. Then, instruction can focus on purposes of discourse common to accounting. Decision Style and Writing Style Generalizations about relationships between writing styles and decision styles on the basis of analyses of memoranda from accountants whose decision styles represent only four of the twelve possible combinations of dominant and back-up styles must be limited. Before we can recommend that businesses administer decision style inventories to help match readers to writers and manpower to writing tasks, we must analyze many writing samples produced by a wide range of decision stylists and elicited from a wide variety of writing tasks in many business environments. Furthermore, we need to determine if isolating other linguistic features will help us recognize a relationship between writing and decision styles more readily then those features selected for this study. 127 While differences in preferences for linguistic features among decision stylists are small, there are sufficient preferences to merit further investigation. One potential outcome of such investigations may be that, as Management Information Systems literature contends, we can match decision style preferences to specific linguistic preferences. (That is not to say, as Cook- Gumperz and Gumperz are quick to note in "From Oral to Written Culture: The Transition to Literacy," "the mere acquisition of discourse strategies will automatically result in social change [the organizational hierarchy being a small social structure].")111 We will then be able to,describe a set of readers for writers and can help them choose strategies that will most efficiently and effec tively communicate with those readers. In addition, while the interrelationship between intended readers and discourse purpose was beyond the scope of this investigation, the findings suggest that that relationship may influence writers' linguistic choices. If that is the case, being able to describe a set of readers by matching decision style preferences (or by any other means) will be useful in describing that relationship. Summary of Study To measure written discourse effectiveness, we must learn what makes certain language choices effective or ineffective for a given purpose and context. First, however, we must describe discourse generated in specific contexts, for specified purposes, and specific 128 readers. To this end, this study describes style in 67 internal memoranda written by 7 junior-level accountants in a Big Eight accounting firm. Influenced by the work of Louis T. Milic and M. A. K. Halliday and Ruqaiya Hasan and insights from sociolinguistics, the discourse analysis framework developed for the study identifies preferences for selected cohesive devices (reference items, verbs and voices, grammatical subjects), elements of coherence (conjunc tions and sentence openers), and elements from which reader-writer relationships can be inferred (hedges, modals, pronouns of address, formulary expressions.) Writers' linguistic choices were examined in relation to (1) their decision style preferences, identified through Rowe's Decision Style Inventory; (2) purposes of discourse, identified by three trained raters; and (3) directions of communi cation, identified by the writers. Results. Choices and frequencies of cohesive devices were more strongly influenced by the three variables than were elements of coherence and items that imply reader-writer relationships. Writers' decision styles influenced linguistic choices least. Although only four of twelve combinations of dominant and back-up decision styles were represented in the seven accountants, linguistic choices of Analytic-Directive stylists differed substantially from those of other stylists, suggesting a relationship between decision and writing styles. Patterns in stylistic choices were apparent in memoranda classified by purpose. As expected, recommendations and requests 129 shared the most features; over half of the total reader-writer relationship items appeared in recommendations and requests. Direction of communication also influenced accountants' linguistic choices, with obvious differences among grammatical subjects of main clauses. Recommended Research. Further research to answer these general questions is recommended: 1. How do linguistic features differ in discourse written (a) in other professions, (b) for other purposes, (c) for other directions of communication, (d) in other modes, such as letters soliciting clients and auditing reports, and (3) by other decision stylists? 2. To what extent is shared knowledge assumed? How does it influence writers' consideration of audience, purpose, and style? 3. How can we help students focus their purposes and consider their audiences? 130 APPENDIX A U N IV ER S IT Y O F S O U T H E R N C A L I F O R N I A S C H O O L O F B U S I N E S S A D M I N I S T R A T I O N i_OS A N G E L E S . C a l i f o r n i a 9 0 0 8 9 1 4 2 1 D E P A R T M E N T O F B U S I N E S S C O M M U N I C A T IO N March 3, 1982 Subject: XYZ1 s Cooperation in Collecting Memoranda for T JSC Di ssernatlon ^ ---------------------------- Dear Mr. X: Professor J. Dquglas Andrews has told me that when you and he discussed my request for writing samples for my dissertation he told you that I would send you a description of my study and how XYS can contribute to the understanding of accountants' writing strategies and learn what kind of writing skills training would most effectively improve written communications within XYS. Purpose of Study. The objectives of my doctoral study, "The Relationship Between Decision-Making Strategies and Writing Strategies of Selected Accountants," are to describe stylistic features in memoranda that fulfill a variety of tasks and were written by accountants whose decision-making styles have been identified as analytic, conceptual, directive, and behavioral (Please see the attached Decision Style Inventory, designed by Alan J. Rowe of USC's Graduate School of Business Administration.), and to identify stylistic preferences of four groups of accountants so that a manager can adapt his writing styl^ to the preferences of his reader, that is, to the style that the reader will process most efficiently. Benefits of the Study to the Accounting Field and to XYZ. The literature about writing in accounting firms, m addition to emphasizing how essential and economical effective communications within a firm are, suggests that the teaching of writing in college accounting, programs is inadequate. From research in composing in general, we can infer that we do not know enough about what users of accounting information need and prefer. My study will contribute to our understanding of accountants' writing strategies and of the stylistic preferences of readers of accounting communications. When the study is completed, XY2 will receive • an overall description of decision-making styles,- • positive and negative writing tendencies noted in the memoranda; 131 • an analysis of each of the accountant's decision making styles; , • a description of the kind of writing skills training that would most effectively improve written commu- ; nication within the firm. . Procedure for XYZ Participation in Collecting Memoranda. As the attached cover letter to the participating accountants notes, two forms are to be completed: • A Fact Sheet; • Alan J. Rowe's brief Decision Style Inventory. The Fact Sheet, which follows the cover letter here, can of course be amended to elicit additional information that would .be helpful to XYZ. The Decision Style Inventory, also at tached, will identify the accountants' decision styles. To save time, the accountants need not tally the scores on the , Decision Style Inventory. |After I have identified the decision styles of the accountants, I will select a number from each category. From them I will request ten memoranda. Five randomly selected memoranda from ■each of these sets will be analyzed. "As Professor Andrews assured you, I will clear with you any reference to XYZ made in my dissertation. Timetable for collecting Memoranda. The Fact Sheet and Decision Style Ihventory--which should take about fifteen minutes to complete— can be submitted to the participating accountants at your earliest convenience. Perhaps, I could .collect the memoranda from your accountants about two weeks later. A progress report on my analysis of the stylistic characteris tics of the memoranda will be submitted to you in late May, and the final report will be ready in early fall. I can be reached at T JSC's Freshman Writing Program Office (743-5672) .on Mondays and Wednesdays and at home on the other weekdays (837-1598). I look forward to your comments regarding an appropriate timetable for you. Sincerely, Betty ?. Pytlik Doctoral Candidate, Department of English cc: J. Douglas Andrews 132 APPENDIX 3 U N IV E R SIT Y O F S O U T H E R N C A L I F O R N I A S C H O O L O F B U S I N E S S A D M IN IS T R A T IO N Los A n c e l e s , C a l i f o r n i a 9 0 0 8 9 - 1 4 2 1 D E P A R T M E N T O F 'B U S I N E S S C O M M U N IC A T IO N March 8, 1982 Dear Accountants: You have probably read recent articles in The Journal of Accountancy and Management Accounting and statements from the American Accounting Association that attribute the success or failure of accounting operations largely to how managers and other users judge written communications of accountants. Your answers to the questions on the attached fact sheet and Decision Style Inventory will help describe the needs of those readers. Specifically, your answers will help determine what relationship exists between decision-making strategies and writing styles of accountants. I will be using the information for my doctoral study at the University of Southern California under the guidance of Professor J. Douglas Andrews, Chairman, Department of Business Communication, and Professors W. Ross Winterowa (Dissertation Chairman) and Betty Bamberg, Department of English. When my study is completed, you and XYZ will receive an overall description of decision-making styles, a description of your individual writing tendencies, and descriptions of the kinds of writing skills that would most effectively improve written communication within XYZ. • Please identify yourself by social security number on both forms. I will use no names in my study. • Please return the completed Pact Sheet and Decision Style Inventory to Mr. X by March 12. To save you time, I will tally the Decision Style Inventory. Sincerely, Betty P. Pytlik Doctoral Candidate Department of English The University of Southern California Attachments, 2 133 APPENDIX C INSTRUCTIONS FOR COMPLETING THE DECISION STYLE QUESTIONNAIRE 1. . FILL IN EACH BOX THAT FOLLOWS EACH QUESTION ON A GIVEN ' LINE WITH THE FOLLOWING NUMBERS: a : MOST LIKE YOU 4 : MODERATELY LIKE YOU 2 : LITTLE LIKE YOU 1 : LEAST LIKE YOU 2. YOU MUST USE ONLY THESE FOUR NUMBERS. 3. YOU CAN'T DUPLICATE NUMBERS IN A GIVEN LINE. USE EACH NUMBER ONLY ONCE. 4. FOR EXAMPLE, THE SEQUENCE IN ANSWERING A QUESTION MIGHT APPEAR AS: 5. ANSWER THE QUESTION FOR THE MOST FREQUENTLY OCCURRING SITUATIONS IN YOUR WORK ENVIRONMENT. S. GENERALLY, THE FIRST THING THAT COMES TO MIND SHOULD BE FILLED IN. 7. THERE ARE NO RIGHT OR WRONG ANSWERS. THE RESPONSES REFLECT' YOUR- PREFERENCES OR REACTIONS TO GIVEN SITUATIONS. 8. THERE IS NO TIME LIMIT IN ANSWERING THE QUESTIONS. 134 SOCIA L SEC U R ITY NUM BER I j T DECISION STYLE INVENTORY III Please score the following questions based on the instructions given. Your score reflects how yo u see y o u rse lf, n o t w n a t yo u believe is correc t or desirable, as related to y o u r w o rk situ a tio n . 1. Mv o rim e ob tectiv e is to; Have a p o sitio n ; w ith sta tu s t Be the b e st in m y field A chieve reco g n itio n : fo r m y w ork \ Feel secure in job j 2. 1 en io y jobs th at: A re tech n ical & j w ell d e fin e d 1 Have co n sid erab le variety A ilow in d eo e n d e n t j actio n i Involve p eo p le ! 3. 1 e x p e c t p e o p le w o n tin g fo r m e to tie: P ro d u ctiv e ; an d fast , i H ighly c ap ab le j C o m m itted a n d ‘ responsive I R ecep tiv e to j suggestions , 4. in m y job, 1 look for: Practical results j T he b est so lu tio n s [ N ew a p p ro ach es o r ' ideas G o o d w orking co n d itio n s l 5. 1 c o m m u n ic a te b est w ith o th ers: O rally and d ire c t r j In w ritin g | By having a ' discussion ! In a gro u p m eeting \ 6. In m y o lan m n g 1 em pnasize: C u rren t n eed s j i M eeting ob jectiv es j F u tu re goals | O rganizational needs 7. W hen faced w ith soiving a o ro b le m . 1; R ely o n proven a p p ro a c h j A p o ly c arefu l ! analysis i L ook for creative approach R ely o n m y ! feelings : 8 . W hen using in fo rm a tio n t orefer: S p ecific facts A c c u ra te a n d j co m p lete d a ta 1 B road coverage of m any o p tio n s L im ited d a ta easily u n d e rsto o d 9 . W hen i am u n c e rta in a b o u t w h a t t o d o . f: R ely on h u n c h i an d in tu itio n i Search fo r facts ! i E xplore a possible i co m p ro m ise j D elay m ak in g a ' i decision J 10. W henever possible. 1 avoid: Long d e b a te s | In co m p lete w ork Using n u m b ers or form ulas i C o n flict w ith I o th ers ! I t . 1 am especially goo d at: R em em b erin g d ates | S » facts Solving d iffic u lt pro b lem s Seeing m any 1 p o ssibilities ; In te rac tin g w ith j o m ers j 12. W hen tim e is im p o rta n t, 1: O e c io e a n d a c t 1 quiCKty Fo llo w o lan s and p rio rities ° e fu s e to be p ressured Seek guidance or su o o o rt ' 13. in social settin g s i generally: S p eak w ith o th ers T h in k a b o u t w n a t \ i? bem g said f Observe w h a t is going on Listen to the co n v ersatio n 14. t am a o o d at rem em b erin g p e o p le 's: N am es ' ■ Places w e m e t 1 Faces P erso n ality ; 15. T h e w o rk 1 d o pro v id es m e: T h e p o w e r to j in flu en ce o th ers I C hallenging assignm ents i A chieving m y 1 perso n al goals ] A c cep tan ce bv ■ the g ro u p i 16. 1 w o rk w ell w ith th o se w h o are: E n erg etic & j a m b itio u s i P u n ctu al & c o n fid e n t 1 C urious & open m in d e d 1 P o lite St tru stin g ■ 17. W hen u n d e r stress, I : B ecom e a n x io u s j C o n c e n tra te o n j the o ro b lem B ecom e fru stra te d \ o r a n n o y e d i Am c o n c ern e d j o r forg etfu l 18. O th e rs co n sid e r me: Aggressive & 1 d o m in e e rin g ! D isciplined & ! precise • Im aginative 8e ! p e rfe c tio n ist 1 $ u o o o rtiv e & co m o assio n ate 19. M y decisions are: R ealistic 3i j ■m oersonal S y stem atic ! a b stra ct : B road an d flexible i Sensitive to the . n eeds o f o tn ers 20. I dislike; N o t b em g in * c o n tro l 1 3 o rin g w ork Follow ing ruies Being reie c te a ; 1 ! 135 APP3JDIX D Betty P. Pytlik University of March 8, 1982 FACT SHEET FOR XYZ ACCOUNTANTS SSN ____________ Age: Sex: M F T How many years have you been in the accounting field, excluding college accounting program? 1 How many years have you been with XYZ? How many memoranda do you write in the Los Angeles office during an average week? . . How many of these weekly memoranda are directed to staff above you? To staff under you? ' To staff at your level? To the file? Other? (Please specify task) How many memoranda do you write to the staff in other XYZ offices in an average week? u q w rnsnv Tnsrrcirs.ncis signatures during an average week? How many courses in written communication for business did you have in college? What training courses in written communication have you had on the job? Thank you very much. Please return the FACT SHEET and DECISION STYLE INVENTORY to ' ______ _________ 136 APPENDIX E U N IV ER S IT Y O F S O U T H E R N C A L I F O R N I A S C H O O L O F B U S I N E S S A D M IN IS T R A T IO N L o s A n g e l e s . C a l i f o r n i a 9 0 0 8 9 - 1 4 2 1 D E P A R T M E N T O F B U S I N E S S C O M M U N I C A T I O N June 24, 1982 Dear Accountant: Thank you for completing the Decision Style Inventory and fact sheet that Mr. X gave to you on my behalf in March. From your responses, I have arrived at a list of accoun tants whose writing habits and decision style preferences represent a rangje of writing styles and tasks. To help me determine what relationships exit between decision- and writing styles of. accountants, may I pleasie --- have additional information from you in the form of internal communications that you have written while— at-— -- XYZ. Specifically, an assortment of 12 to 15 internal communications (memos, short and long reports— in short, anything except working papers) written to various readers and for various purposes would be most usefu-l-to-- - ■ my study of stylistic features. In the upper right corner of each communication, please write vour social security number. In addition to that -identification,- ! need to- know the direction and purpose- .. of each communication. Using the ierrsrs and numbers below, please mark in the upper right corner the direc tion and the main purpose of the communication. - -- The Direction of the Communication A. To staff above you. B. To staff under you. C. To staff at your level. - D. To;the file :................ E. To other (Please specify). 137 XYZ Company -2- June 24, 1982 Accountant Main Goal of Internal Communication When, for example, a communication recommends an action and at the same time requests further information, please choose the goal that you think is the more pressing one. 1. To Describe. (e.g., to describe a procedure, ' which .may also entail classifying, identifying) -- ' 2.1 To Recommend. (To propose, direct, order) ' . 3. "To Request. (To' ask, inquire) 4. ' To Document. (To verify) ' ' ~ Although I assure you that XYZ will not be mentioned in , my study, nor will your written communications be reproduced in . the -study,-- please-delete- information • "that* • you deem confidential. ..... I know that this is not a'five-minute task I have asked = you to complete, and I am -most- appreciative of your cooperation. When my' study is completed', you will receive ' an overall description ' of decision ' making style's, _ a' description of.' 'your individual writing' tendencies and descriptions ' of the kinds of writing skills that may increase you writing productivity. Please give the 12 to 15 communications to Ms. Y by Friday, July 2. . Sincerely, ...... Betty P. Pytlik Doctoral Candidate Department of English'......... The University of Southern California 138 APPENDIX E FACTS ABOUT THE PARTICIPATING ACCOUNTANTS The following information about - the' - seven - accountants who provided the writing samples for this study was obtained from the fact sheets completed by the accountants. Accountant j 1, a 30-year-old female, has been in the accounting field for three years and with the participating firm for two years. She 'estimated - that' she writes five memoranda per week,- • all to -the file.' She has had neither college courses nor in-house - training in written business - communication." Her decisibn^st'yTeJ±s-Anaiytfh- ;'her back:-- up style is Directive. Accountant 2, a 32-year-old male, -has worked only for ^ 2 ^ 1 3 ^2c i i c i . ~ i . r i c j S i z r m e l i x i r o y i s s y j ‘2 < s n 2.1*1 accountant. He estimated that he writes ten‘memoranda per week, six to staff above him, one each to staff under him ..and at his own level, and two to clients. Some memoranda go to other offices in the firm; some are written for others' signatures. Although he has had no in-house training in writing, he had a college course in written business commu nication. His -decision style preference is Analytic; his back-up style is Directive. Accountant 3, a 35-year-old male who has been with the participating firm for the eight years he has been an 139 accountant, estimated he writes ten memoranda per week, seven to clients and others, one to staff above him, two to staff below. One memoranda is written for another staff member's signature. He has had one college business commu nication course and no in-house training in writing. His decision style preference is Analytic; his back-up style is Conceptual. Accountant 4, a 27-year-old male, has worked for the participating fjirm for the four and one-half years that he has been an accountant. The one memorandum that he writes each week is either to ''document audit planning" or to "inform other offices [in the firm] of inventories, etc." It is written for another staff member's signature. ' He has had one 'college course In'writing for business, but he has had no in-house training. His decision style preference is Analytic; his back-up style is Conceptual. Accountant 5, a 24-year-old first-year accountant, estimates that he writes three memoranda per week, two to staff above him, one to staff below. One of these is written for another staff member's signature. He had a college course in writing for business and was about to take the firm's course in effective written communication. Kis dominant decision style is Conceptual; his back-up style is Directive. Accountant 6, a 29-year-cld male with six years of accounting experience, all with the participating firm, estimated that he writes five memoranda per week, three to 140 clients and two to staff above him. One is written for another's signature' one is sent to another ofr the firm' s- - offices. He took no business communication courses ; in college, but had an in-house training course. His--decision style -preference is Conceptual; his back-up style . -is Analytic. Accountant 7, a 34-year-old male who has been with the participating firm, -for the six years :he has been an -accoun tant, estimated thajt he writes ten memoranda per week,; one - for another's signature, and one for -staff - . in another, office., . Of . the., ten . . memoranda, . two are directed, to. staff.- above him,:-two. to staffs under, hi-m-,- - one-.-fo staff.- at : the , same- 7. : . level,, and five to the file. He took two college courses in writing for business and completed an; in-house -effective written, communication course. His. decision - . style preference . is Analytic; his.back-up style is.Conceptual.' 141 Appendix G __ _ ILLUSTRATIONS OF THE FIVE CATEGORIES OF MEMORANDA PURPOSES Three trained readers and the writers agreed on the purposes of the following memoranda. Memorandum 311 1. Description. Subject: All Counseling Sessions/P-88 Preoaration i , . . ' Attached are the a’ nr.ual time journals for the staff members that you counsel. These journals will help.you identify .the engagements for which your counselees should have received P-66's. If a personnel file^does not contain all the P-66's due an individual, the managers/supervisors of the engagements for which. ; P-66' s were net received should be notified immediately. Memorandum 202- - - ■ r. ' 2. Recommendation- : ; , l : . ■Subject: Scheduling .. The following PSAS staffing schedules are enclosed for your use: • Completed.schedule for July through September 1982 • 31ank schedules for October 19.82 through March 1983. • Prior year's schedules for October 1981 through March 1982. Please complete the blank schedules for all of your staffing requirements through March 1983 and return them to me bv July 16, 1932. 142 Memorandum 707 3. Request ... .... X. Subject: Job Opening at Prospective Client XYZ Co. is proposing - to obtain the audit services of ABC, Inc. I just learned that they are interested in hiring an accountant for their corporate staff, and I feel if we can respond quickly with a potential candidate, our proposal efforts may be aided. If you are aware of anyone, within XYZ Co. or elsewhere, who meets the qualifications below and is interested, please call me at once. ABC, Inc. 's requirements are: • Bachelojr degree in accounting • Familiarity . with , the . MSA .data.processing .general . ledger system • One to. two years experiencejin public, accounting . or private industry.with MSA system :.... The position initially will pay $23,000 - $25,000 and can lead to internal audit and financial analyst positions. The position reports, to the Director of Financial Services, who in turn reports tc the Controller. Please advise me or my secretary Y, in the XXX office, by Thursday afternoon-January 7, if you know of any prospects. Memorandum 603 4. Documentation Subject: .ABC Co. - Review of Prior Auditors' Workpapers Following is a summary of items requiring planning consideration as noted during review of DEF & Co.:workpapers for their September 30, 1981 audit examination of ABC Co., Inc. : • Significant Adjusting Journal Entries $480 - samples recorded as sale/receivable $7,866 - unrecorded credit memos $28,521 - accounts payable accrual $3,406 - accrual of sales commission 143 Accounts Receivable $749 - passed adjustment for subsidiary ledger balance in excess of G/L Sales cutoff testwork not complete. Items selected from October invoices and not ' resolved implied gdods were shipped in " September. Of 38 invoices selected, 5 were not traced to shipping documentation (i.e., UPS log). No accrual was recorded "for estimated sales returns'. Returns subsecruent to September 30, 1981 and March 31; 1982 will be examined to de-termine if an accrual is warranted. Finished Goods Piece goods out at printers of $5,933. Properly confirmed. . - price testwdrk appears adequate': - _ Purchase cutoff testwork appears adequate. Other Assets' ” ' • - - - - ..... Employee and contractor loans receivable totaling $3, 245 were not confirmed or agreed to supporting documentation. Deposits of $2,675 were noted as being agreed to "cash disbursement." No mention of cancelled check. Prepaid expenses include $7,375 of product samples that have been sent to potential customers. No support for amount of samples. Capitalisation is questionable as these samples would only appear to benefit the period in which they are distributed. Capitalized trademark expense of $3,420 represents several invoices for company logo design and box fronts and is being amortized over five years. Invoices were not vouched. Details need to be reviewed with X as some initial designs may have.been abandoned with the recent display'and box "changes . ~ ^ Capitalized patent expense of $8,685 represents cost of securing patent rights for product and is being amortized over 17 years. 144 Invoices were not vouched; 17-year useful life needs to be addressed. Capitalized catalog' costs of $13,513 are amortized over a.three-year life and were not vouched to cancelled checks (only partially agreed to cash disbursements journal). As Company has just recently redesigned promotional catalogs,- future benefit of these costs will need to be addressed. Accounts Payable Confirmation effort -and search for unrecorded liabilities appear adequate. Slj, 860 -was noted as being payable to MNO as the result of a court judgment. This item was not included as a liability at September 30, 1981, although it relates to a prior period and was paid in the subsequent period. Other Matters Attorney letter from P & 0 was not requested due to DEF1 s representation that only XXX matters were being handled. There is no indication that this representation was verified by review of related invoices. Following is an example of ‘a "5” memorandum. Two readers identified it as a 4 (Documentation); one identified it as a 1 (Description). The writer identified it as having two purposes. Memorandum 602 5. Two or More Purposes Subject: Response to Explanation of Negative Responses - 2 Concerning Self-Review of June 30, 1980 Audit Examination-of ABC Co. Question 146 - Question addresses whether substantive samples are representative .of their respective populations. Departure is classified as minor as only account 145 receivable balances greater than $20,000 were selected for confirmation. Response: Accounts receivable confirmation scope resulted, in th selection of account balances aggregating -72% of the dolia value of accounts receivable at May 31, 1980. -The accoun balances not selected for confirmation consisted of numerous, individually insignificant items and additional sampling was not considered necessary as the likelihood of a material error was considered small. 147 ' - Question asks if all items selected for - testing : were audited to a conclusion. Departure is classified as minor as alternate fcestwork for non-responding accounts receivable confirmations was not considered complete. Response: Of the $2,128,045 in confirmation requests requiring alternate procedures, $1,855,889 or 87% were vouched to subsequent cash receipts. As the remaining balance of accounts requiring alternate testwork was minor ($272,157) and based on analytical review of ; the remainder which indicated the majority were May sales (i.e. reasonable that amounts were not collected as of last day of field work), additional testwork was passed. 150 - Question addresses the propriety of the aucit conclusion m relation to tne possm^e aggregate error projected from the sample. Departure classified as minor and relates to accounts receivable confirmations. Response: Question is N/A as no exceptions ' were noted during our evaluation of the-validity of accounts receivable balances. r f ^ <P 146 Appendix H Instructions to the Memoranda Readers Betty P . Pytlik July 16, 1982 Instructions for Identifying the Main Purposes of Memoranda . to - 3e Used. in My Dissertation Study. I asked accountants at the Los Angeles office of a major accounting firm to select from a short list of communication purposes the main purpose of each memoranda they gave me for my study of style-. Here are my instructions to the accountants: Main Goal oi' Internal Communication When, for example, a communication recommends an action and at the same time requests information, please choose the goal that you think is the more pressing one. : 1. To Describe. (e.g., to describe a procedure, which may.also entail classifying, identifying) 2. To Recommend. (To propose, direct, order) 3. To Document. (To verify) Using the same choices (and numbers), please indicate in the appropriate spaces on- the attached scoring sheet what you believe to be the main goal of each of the attached memos. Thank you for your help. You will, of course, be hearing from me periodically regarding what I have learned about stylistic features of internal memos. Please call me in the evenings or. on week-ends if you have any questions; 450-1612 is my number. CAMPUS ) MEMO FROM DATE: 147 APPENDIX I EXAMPLE OF CODED MEMORANDUM Memorandum 212 Enclosed are all working papers related to the procedures you requested we perform at ABC Co. ^ I understand that X has told you that our audit at ABC Co. for the year ended January 31, 1982 is not yet complete. At this time, we believe there will be an adjustment related to accounts receivable that may affect the - December 31, 1981 balances. Also, we have made an adjustment of approximately $1,200,000, reducing inventory and accounts payable at January 31, 1982. The client has not yet provided the data needed to determine if this entry affects the December 31, 1981 balances. ; We will inform you of any changes to the December 31, 1981 balances as soon as the information is available. Hedges Modals 2 Rounders 1 Negative 1 will/would Conjunctions 1 Additive Ooeners Grammatical Sub~j ects Kinds of Verbs and Voices 1 Prepositional Phrase 1 Adverb 1 Inverted Phrase 5 Names 1 Other 1 Passive Voice 5 Active Voices as-soon as the information is available approximately $1,200,000 has not yet provided will inform you Also At this time A1 so Enclosed are .... I, we, we, client, we Working papers Enclosed are .... I understand that .... We believe (that) .... We have made an adjustment. . The client has not yet provided the data .... We will inform you .... 148 i References 8 Personal You, we, I, you, we, we, Pronouns we, you 1 Possessive Our Determiner 8 the the procedures the year ended January 31, 1982 the December 31, 1981 balances the client the data the December 31, 1981 balances the December 31, 1981 \ balances \ X the information 2 Demonstra this time tives this entry - v - 149 NOTES - 1 Donna Stine and Donald Skarzenski, "Priorities for the Busi ness Communication Classroom: A Survey of Business and Academe," Journal of Business Communication, 16 (Spring, 1979), 15-30; Paul V. Anderson, Research into the Amount, Importance, and Kinds of Writing Performed on the Job by Graduates of Seven University Departments That Send Students to Technical Writing Courses, paper presented at the annual meeting of the Modern Language Association, Houston, December 1980; and Lester Faigley, Thomas P. Miller, Paul R. Meyer, and Stephen Witte. Writing After College: A Stratified Survey of the Writing of College-Trained People. Writing Program Assessment. Technical Report Number 1, 1981 ED 210. 708. 2J. Douglas Andrews and R. J. Koester, "Communication Difficul ties as Perceived by the Accounting Profession and Professors of Accounting," Journal of Business Communication, 16 (1979), 33-42. 3Dixie Goswami makes a case for needed research in nonacademic discourse in "Naturalistic Studies of Nonacademic Writing," Moving Between Practice and Research in Writing: Proceedings of the NIE-FIPSE Grantee Workshop," ed. Ann Humes with Bruce Cronnell, Joseph Lawlor, Larry Gentry (Los Alamitos, California: SWRL Educational Research and Development, 1981), pp. 143-147. With Lee Odell, she has herself contributed significantly to our understanding of nonacademic discourse in their 1981 Fund for the Improvement of Post-Secondary Education study, Writing in Non-Academic Settings. ^Among the discourse theorists who have urged greater attention to how purpose and context determine stylistic choices are Francis Christensen in Notes Toward a New Rhetoric: Six Essays for Teachers (New York: Harper and Row, 1967); Robert de Beaugrande in "Psychology and Composition," CCC, 30 (1979), 50-57; C. H. Knoblauch in "Intentionality in the Writing Process: A Case Study," CCC, 31 (1980), 160-168; and Joseph M. Williams in "Non-Linguistic Linguistics and the Teaching of Style," Linguistics, Stylistics, and the Teaching of Composition, ed. Donald McQuade (Akron, Ohio: Department of English, University of Akron, 1979), pp. 24-40. SA Quantitative Approach to the Style of Jonathan Swift (The Hague: Mouton & Co., 1967). 150 6Milic, p. 75. 7Milic, p. 17. 8(New York: Teachers College Press, 1977), p. 135. 9In Essays on the Language of Literature, ed. Seymour Chatman and Samuel Levin (Boston: Houghton Mifflin, 1967). In his "Criteria for Style Analysis" (in Essays on the Language of Literature, pp. 412-430), Michael Riffaterre, like Milic and Hiatt, criticizes subjective impressionism, normative rhetoric, and premature aesthetic evaluation and advocates a linguistic structure description of style. His distinction between description and stylistic description is pertinent to this study: "A purely linguistic analysis of a work of literature will yield only linguistic elements; it will describe those elements of a sequence which happen to have a stylistic value along with the neutral ones; it will isolate no more than their linguistic functions without discerning which of their features make them stylistic units as well. Still, the application of linguistic methods to such units would give us an objective knowledge of their double role as elements of both the linguistic and the stylistic systems. But this application requires a preliminary sorting" (p. 412). 1°Halliday, p. 218. 11Parts 1 and 2 appeared in 3 (1967), pp. 37-81, 19-244; Part 3 in 4 (1968), pp. 179-216. 12"The Case for Case," in Universals in Linguistic Theory, ed. Emmon Bach and Robert T. Harms (New York: Holt, Rinehart and Winston, 1968), pp. 1-88. 13Staging occurs when we make one element of what we are discussing the point of departure for the relationships to all other elements. See "staging" and "thematization" in Part 2 of Halliday's "Notes" and in Joseph E. Grimes' The Thread of Discourse (The Hague: Mouton Publishers, 1975). Information coherence, also discussed in Part 2 of "Notes," occurs when "the speaker organizes the components of the message block in such a way as to specify its relation to what has preceded" (p. 20). 1<‘The relationship that the patient has with the verb in a clause is that it is the thing or person that is changed in form by a process. For further discussion, see Grimes, especially pages 123-124. Halliday also refers to "focus" in connection with work order. Focus "involves the selection, within each information unit, of a certain element or elements as points of prominence 151 within the message" ("Notes: Part 2," p. 203). Wallace Chafe refers to the same phenomenon as "topic." See "Givenness, Contrastiveness, Definiteness, Subjects. Topics, and Point of View" in Subject and Topic, ed. Charles Li (New York: Academic Press, 1976), pp. 27-25. 15(London: Longman, 1976), p. 4. ieHalliday and Hasan, p. 4. 17Halliday and Hasan, p. 31. 18In Style in Language, ed. T. A. Sebeok (Cambridge, Mass.: M.I.T. Press, 1960), p. 214. 19"Style and Its Consequences: Do As I Do, Not As I Say," CE, 43 (1981), 433-451. 20See also Williams' text, Style: Ten Lessons in Clarity and Grace (Glenview, Illinois: Scott, Foresman, 1981), for a discussion and examples of occasions when nominal forms are preferable to verbal forms. Veda Charrow and Flower, Hayes, and Swarts also discuss nominal versus verbal forms in Let the Rewriter Beware and Revising Functional Documents: The Scenario Principle, respectively. 21Wells, pp. 217-218. 22Hake and Williams, p. 436. 23In Subject and Topic, p. 27. 2<lGrimes, p. 97. 2SHerbert H. Clark and Eve V. Clark summarize the research on processing passives in Psychology and Language: An Introduction to Psycholinguistics (New York: Harcourt Brace Jovanovich, Inc. , 1977). 26Again, Williams' Style: Ten Lessons in Clarity and Grace is useful in assessing the potential of passives and in describing how they function in discourse. 27See Clark and Clark for a summary of the studies relevant to processing prose with agents as subjects. An agent, according to Joseph Williams, is "the originating force of an action, the source of an action, the responsible party, that entity without which an action should not occur" (p. 96). See Style: Ten Lessons in Clarity and Grace. 152 2®Technical Report No. 10. (Pittsburgh, Pennsylvania: Carnegie-Mellon University, 1980). 29(0xford: Oxford University Press, 1978), p. 52. 3°Widdowson, p. 44. Conventions such as genres and Grice's rules of conversation. First published in "Logic and Conversation," in Syntax and Semantics, Vol. 3: Speech Acts, ed. P. Cole and J. L. Morgan (New York: Academic Press, 1975), pp. 41-58, Grice's maxims are: 1. Quantity: Be as informative as required. Be no more informative than required; 2. Quality: Say only what you believe to be true; 3. Relevance: Be relevant; 4. Manner: Be perspicuous. Don't be ambiguous. Don't be obscure. Be succinct. 31Halliday and Hasan, p. 227. Several descriptive procedures for representing propositions in discourse have been developed by text linguists like Walter Kintsch, Teun van Dijk, and Edward J. Crothers and theorists of style like Louis T. Milic. These procedures rely heavily on intuition and are more time consuming and rigorous than this study required. Kintsch discusses connections of propositions in The Representation of Meaning in Memory (Hillsdale: Lawrence Erlbaum Associates, 1974); van Dijk in Text and Context (London: Longman, 1977); Crothers in Paragraph Structure Inference (New Jersey: Ablex Publishing Corporation, 1979); and Milic in Stylists on Style (New York: Charles Scribner's Sons, 1969). In "The Grammar of Coherence," in Contemporary Rhetoric: A Conceptual Background with Readings (New York: Harcourt Brace Jovanovich, 1975), pp. 225-233, W. Ross Winterowd claims that "beyond the sentence marker . . . we perceive coherence only as the consistent relationships among transitions." The relationships-- Winterowd identifies six--"constitute the grammar of coherence for all units of discourse beyond the level of what [Winterowd calls] syntax" (p. 228). The words and phrases that Winterowd calls transitions (as does Lyman Hagen in "An Analysis of Transitional Devices in Student Writing,"- RTE, 5 [1971], 190-201) are variously called "sentence connectors" (by Nancy Arapoff in "The Semantic Role of Sentence Connectors in Extra-Sentence Logical Relationships," TESOL Quarterly, 2 [1968], 243-253), "junctions" by Robert de Beaugrande in Text, Discourse, and Process: Volume IV--Advances in Discourse Processes (New Jersey: Ablex Publishing Company, 1980), and "conjunctions" by Halliday and Hasan. Arapoff offers a classification of five binary relationships: inferential, comparative, additive, intensifying, and exemplifying. Hagen provides another classification of transitions, categories identified according to purpose: to introduce an illustration, to add another phase of the same idea, to affirm, to indicate a conclusion or result, to point to a contrast or qualification, to concede, to introduce, to show cause or to indicate a relationship 153 in time, to show purpose, to restrict, to repeat, to show time and place. In adopting their scheme of four categories of semantic relations (additive, adversative, causal, and temporal), Halliday and Hasan explain that "There is no single, uniquely correct inven tory of the types of conjunctive relations; different classifica tions are possible, each of which would high-light different aspects of the facts" (p. 238). They contrast conjunctions with the other categories of cohesive devices--reference, substitution, ellipsis, and lexical. While the devices in the other four categories give the readers direction for interpreting an element in terms of its textual environment, conjunctions don't provide search instruction but instead specify "the way in which what is to follow is systema tically connected to what has gone before" (p. 227). 3 2Notes Toward a New Rhetoric: Six Essays For Teachers (New York: Harper and Row, 1967), p.44. 33Alan J. Rowe, "Decision Making in the '80's," Los Angeles Business and Economics, (1980), p. 7. 3‘ ‘Rowe, p. 7. 3SMichael Driver and Alan J. Rowe, "Decision-Making Styles: A New Approach to Management Decision Making," in Behavioral Problems in Organizations, ed. C. Cooper (Englewood Cliffs, New Jersey: Prentice-Hal1, 1980), p. 178. See also Alan J. Rowe, Richard 0. Mason, and Karl E. Dickel's Strategic Management and Planning: A Methodological Approach (Boston, Mass,: Addison Wesley, 1982). 36Driver describes his model in several sources. "individual Decision Making and Creativity" in Organizational Behavior, ed. S. Kerr (Columbus, Ohio: Grid Publishing, 1979) is especially informative because Driver defines decision styles and describes the important decision style models. 3’Michael Driver and T. Mock, "information Processing, Decision Style Theory and Accounting Information Systems," Accounting Review, 50 (1975), p. 497. 3 8Zmud, p. 1089. 39Michael Driver, Administrative Problem Solving Exercise (IST-RII) , 1971. Driver's model and decision-style inventory have been used in connection with written communication: J. Douglas Andrews, Chairman of the Business Communication Department in the Graduate School of Business Administration, the University of Southern California, has administered the inventory to his business communication students to help them analyze "audiences to which 154 accountants must write or to which accountants must speak. By isolating those elements critical to the decision making process, the student can gain an understanding of what information an individual needs to be able to make a decision." Andrews described their uses in a paper presented to the American Business Communication Association Convention, October 16, 1981: "Audience Analysis: A New Application of Decision Style Theory: A Case Study in Accounting Communication." ""Decision Style Inventory, 1981. "’"The Rhetorical Situation," Philosophy and Rhetoric, 1 (1968), pp.' 1-14. "2From the directions for Decision Style Inventory III. 1,3"Decision Making in the '80's," p. 7. ""From "Decision Making in the '80's," p. 8. "5See pp. 402-405 for complete descriptions of the decision styles. "6Preliminary to his doctoral study "Unstructured Decision Making in Strategic Planning" (University of Southern California, 1982), Richard Mann compared Alan J. Rowe's inventory to seven cognitive style measurements and found Rowe's inventory to be the most discriminating in relation to decision styles. (Personal interview with Richard Mann, 8 June 1982.) "’"Defining and Assessing Competence in Writing," in The Nature and Measurement of Competency in English, ed. Charles R. Cooper (Urbana, Illinois: NCTE, 1981), p. 102. "8In his discussion of relative readability in The Philosophy of Composition (Chicago, Illinois: The University of Chicago Press, 1977), E. D. Hirsch reiterates Herbert Spencer's contention that if the writer's purpose is clear to the reader, the reader can devote more time to considering the writer's ideas. "9Hirsch, p. 182. 5“James L. Britton, T. Burgess, N. Martin, A. McLeod, and H. Rosen, The Development of Writing Abilities 11-18) (London: Macmillan Education, 1975) . 5’Frank J. D'Angelo, 'A Conceptual Theory of Rhetoric (Cambridge, Mass.: Winthrop Publishers, Inc., 1975). 155 52James L. Kinneavy, A Theory of Discourse (Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1971). 53"Defining and Assessing Competence in Writing," p. 106. 5*"Psychology and Composition," p. 57. 5SAustin's How To Do Things With Words, ed. F. 0. Urmson and Marina Sbisa, 2nd ed. (Cambridge, Mass.: Harvard University Press, 1975). Searle's work: (Cambridge: Cambridge University Press, 1969); Language in Society, 5 (1976), 1-23; Philosophy in America, ed. M. Black (Ithaca, New York: Cornell University Press, 1965), pp. 221-239. For a summary of the applications of speech-act theory to written discourse, see Martin Steinmann, Jr.'s "Speech-Act Theory and Writing," in Martin Nystrand, ed. , What Writers Know: The Language, Process, and Structure of Written Discourse (New York: Academic Press, 1982), pp. 291-323. 56For an excellent summary of the work of major sociolinguists, see Pier Paola Giglioli's introduction to Language and Social Context, ed. Pier Paola Giglioli (London: Penguin Education, 1972), pp. 7-17. 5 7See the several examples of Searle’s work cited in note 55; William Labov, "The Logic of Nonstandard English," Georgetown Monographs in Language and Linguistics, 22 (1969), pp. 1-22, 26-31. Excerpts reprinted in Giglioli's Language and Social Contexts; C. A. Ferguson, "Diglossia," Word, 15 (1959), 325-340, and "The Structures and Use of Politeness Formulas," Language in Society, 5 (1976), 137-152; Dell Hymes, "Toward Ethnographies of Communication: The Analysis of Communicative Events," in Giglioli's Language and Social Contexts, pp. 21-44; H. P. Grice, "Logic and Conversation," in Syntax and Semantics, Vol. 3: Speech Acts, ed. P. Cole and J. L. Morgan (New York: Academic Press, 1975), pp. 41-58; J. McH. Sinclair and R. M. Coulthard, Towards an Analysis of Discourse: The English Used by Teachers (London: Oxford University Press, 1975); and M. A. K. Halliday, Angus McIntosh, and Peter Strevens, The Linguistic Sciences and Language Teaching (London: Longmans, 1964). 58Sociolinguists interested in the speaker's use of questions include W. P. Robinson and Susan J. Rackstraw, A Question of Answers, Vol. 1 (London: Routledge and Kegan Paul, 1972) and Esther N. Goody, "Toward a Theory of Questions," in Questions and Politeness: Strategies in Social Interactions, pp. 17-55. Several who are interested in politeness formulas are Susan Erwin-Tripp, "is Sybil There? The Structure of Some English Directives," Language in Society, 5 (1976), 25-66; Charles A. Ferguson, "The Structure and Use of Politeness Formulas," cited above; and Robin Lakoff, "The Logic of Politeness; or Minding Your 156 P1 s and Q' s," in Papers from the Ninth Regional Meeting of the Chicago Linguistic Society, pp. 292-305. George Lakoff has discussed hedges in "Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts," in Papers from the Eighth Regional Meeting of the Chicago Linguistic Society, pp. 183-228. Robin Lakoff has written on modality in "The Pragmatics of Modality," in Papers from the Eighth Regional Meeting of the Chicago Linguistic Society, pp. 220-246, and in "Language in Context," in Language, 48 (1972), pp. 907-927. Pronouns are dealt with in Barrie Thorne and Nancy Henley's collection of articles, Language and Sex: „ Difference and Dominance (Rowley, Mass.: Newbury House Publications, 1975). 59Pratt, Toward a Speech Act Theory of Literary Discourse (Bloomington: Indiana University Press, 1977); Fish, "How to Do Things with Austin and Searle: Speech Act Theory and Literary Criticism," Modern Language Notes, 91 (1976), 983-1025; Garnica, "Rules of Verbal Interaction and Literary Analysis," Poetics, 6 (1977), 155-168. 60Investigating English Style (London: Longman, 1969). 61The Language of the Law (Boston: Little, Brown and Co.). 62Daniel B. Felker, ed., Document Design: A Review of the Relevant Research (Washington, D. C. : American Institutes for Research, 1980). 63George Lakoff. p. 183. 61*Esther N. Goody discusses hedges in terms of their contexts in "Towards a Theory of Questions," in Questions and Politeness: Strategies in Social Interactions, pp. 17-55. 6 5See note 30. 66Robin Lakoff, "Pragmatics," p. 230. 67"0n Hedging in Physician-Physician Discourse," an unpublished paper, 1980. 68Prince et al., p. 10. 69Prince et al., p. 11. 70In Papers from the Eighth Regional Meeting of the Chicago Linguistic Society (Chicago, Illinois: University of Chicago, 1972), p. 230. 71(London: Longman, 1979). 157 72In Style in Language, ed. T. A. Sebeok (Cambridge, Mass.: M.I.T. Press, 1960), pp. 253-276. 73With Robbin Battison, Dixie Goswami surveys trends in clear writing in "Clear Writing Today," The Journal of Business Communication, 18 (Fall, 1981), pp. 5-16. Linda S. Flower, J. R. Hayes, and H. Swarts also summarize current demands for clear prose in Revising Functional Documents: The Scenario Principle. Technical Report No. 10 (Pittsburgh, Pennsylvania: Carnegie-Mellon University, 1980). 7k(Washington, D. C.: American Institutes for Research, 1980), N. P. 75Flower et al. (Pittsburgh: Carnegie-Mellon University, 1980); Atlas (Pittsburgh: Carnegie-Mellon University, 1979); Rose and Cox (Washington, D. C.: American Institutes for Research, 1980); and Daniel B. Felker, Pickering, R. Charrow, V. Holland, and J. C. Redish (Washington, D. C.: American Institutes for Research, 1981). 76Revising Functional Documents, p. 1. 77Linda Flower and John R. Hayes explain protocol analysis in "A Cognitive Process Theory of Writing," CCC, 32 (1981), 365-387: Protocol analysis captures "a detailed record of what is going on in the writer's mind during the act of composing itself" (p. 368). "To collect a protocol, we give writers a problem, such as 'Write an article on your job for the readers of Seventeen magazine,' and then ask them to compose out loud near an unobtrusive tape recorder. We ask them to work on the task as they normally would--thinking, jotting notes, and writing--except that they must think out loud. They are asked to verbalize everything that goes through their minds as they write, including stray notions, false starts, and incomplete or fragmentary thought. The writers are not asked to engage in any kind of introspection of self-analysis while writing, but simply to think out loud while working like a person talking to herself" (p. 368) . 78Revising Functional Prose, p. 19. 79Atlas, pp. 1-2. *“Atlas, p. 41. 81(Glenview, Illinois: Scott, Foresman, 1981). 82In Linguistics, Stylistics, and The Teaching of Composition, pp. 26-27. 158 83Williams acknowledges the importance of the work of Kellogg Hunt and Roy C. O'Donnell, however. See Hunt's "Syntactic Maturity in Schoolchildren and Adults," Monographs for the Society of Researchers in Child Development, No. 134, Vol. 35, No. 1 (February, 1970) and O'Donnell's "A Critique of Some Indices of Syntactic Maturity," Research in the Teaching of English, Vol. 2, No. 1 (Spring, 1977), pp. 49-53, both of which are cited in Williams' article. 8 ' ' Linguistics, Stylistics, and The Teaching of Style, p 27. 8 5 In What is 'Plain English,' Anyway? (Washington, D. C.: American Institutes for Research, 1979), Veda Charrow discusses the status of our knowledge about plain English: "The fact is, there is a great deal of controversy about the nature of plain English, and one person's plain English may well be another person's gobbledygook. To date, almost no research has been done specifically on the nature of plain English, and only a relatively small amount has been done to determine what kind of written and spoken language is clear and understandable to the average person. We may all think we know what plain English is, but I doubt if any of us could provide a detailed definition with which everyone else would agree. Empirical research must be done to clarify what kinds of grammatical and semantic constructions, vocabulary, and discourse structures are easier or more difficult for people to understand. Otherwise, rewriting will continue to be a hit-or-miss proposition" (p. 8). 86(Washington, D. C.: American Institutes for Research, 1979), p. 1. 87See "Writing in a Non-Academic Setting," Research in the Teaching of English, 16 (1982), pp. 201-223, and Writing in Non-Academic Settings (Washington, D. C.: Fund for the Improvement of Post-Secondary Education, 1981) (ERIC ED 214 163). 88"What Can Evaluators Learn from Writers in Non-Academic Settings?" in Proceedings of Maryland Composition Conference. 16 April 1981, ed. Susan Kleimann and Anne Franzak (College Park, Maryland: University of Maryland, 1982), p. 24. "Described in Hake and Williams' "Style and Its Consequences: Do As I Do, Not As I Say," College English, 43 (1981), pp. 433-451. 90DA, 38 (1978), 7300A-7301A (University of Minnesota). Crowhurst's study is also described in an RTE article co-authored with Gene Piche, "Audience and Mode of Discourse Effect on Syntactic Complexity at Two Grade Levels," 13 (1979), pp. 101-109. 159 91DA, 39 (1979), 4219A (University of Toronto), p. 4219A. 92"A Comparison of Cohesive Elements in American Business and Non-Native Speaker Written Discourse," Diss. University of Southern California, 1979. "DA, 39 (1979), 3997A (Brigham Young University) . 94DA, 42 (1981), 775A-776A (Carnegie-Mellon University) . 95DA, 39 (1979), 7201A (University of Minnesota). Rubin's study is also described in an RTE article co-authored with Gene Piche, "Development in Syntactic and Strategic Aspects of Audience Adaptation Skills in Written Persuasive communication," 13 (1979), pp. 293-316. 96"A Cognitive Process Theory of Writing," CCC, 32 (1981), pp. 365-387. 97(Cambridge, Mass.: The M. I. T. Press, 1975). ""Cognitive Psychology," pp. 27-39. "DA, 42 (1981), 1519A (Rutgers University). 18°The Journal of Business Communication, 19, No. 1 (1982), pp. 55-69. 101Soldow, p. 66. 102DA, 41 (1981), 2898A (Northern Illinois University). Smeltzer also describes his study in "The Relationships Between Writing Style and Leadership Style," The Journal of Business Communication, 18 (Spring, 1981), pp. 23-32. 103DA, 38 (1977), 1833A (The University of Oklahoma) . 104The partner requested that both the firm and the participating accountants remain anonymous. 105In the literature on evaluation of writing, other terms for "trained reader" have been used. In Charles R. Cooper and Lee Odell's Evaluating Writing: Describing, Measuring, Judging (Urbana, Illinois: NCTE, 1977), for example, "assessor," "qualified judge," "evaluator," and "rater" are used. In all cases, as Lloyd-Jones explains in "Primary Trait Scoring" (in the Cooper and Odell collection, pp. 33-66), the reactions of qualified readers are calibrated to a single reporting code. 160 106"Reliability of Content Analysis: The Case of Nominal Scale Coding," The Public Opinion Quarterly (Fall 1955), pp. 321-325. 107The unit of discourse used in this study is the T-unit, "a main clause plus any subordinate clause or nonclausal structure that is attached to or embedded in it." Introduced by Kellogg Hunt in 1964, the T-unit has frequently been used to measure syntactic maturity. Here it is simply employed as a convenient unit for contrasting the frequencies of the selected linguistic features according to writers' decision styles, purposes of discourse, and direction of communication. 108Discussed in "Introduction: The Five Key Terms of Dramatism," A Grammar of Motives (Berkeley: University of California Press, 1945), xv-xxiii. 109"The Rhetorical Situation," Philosophy and Rhetoric, 1 (1968), pp. 1-14. 11“Marilyn Cooper, "Context As Vehicle: Implicatures in Writing," in What Writers Know: The Language, Process, and Structure of Written Discourse, ed. Martin Nystrand (New York: Academic Press, 1982), p. 106. xllJenny Cook-Gumperz and John J. Gumperz, in Variation in Writing. Ed. M. F. Whiteman (Hillsdale, N. J.: Erlbaum, 1981), pp. 89-109. 161 BIBLIOGRAPHY Addams, Harvey L. "Opinions of Practicing Accountants: Written Communication Competencies Necessary in the Accounting Profes sion." DA, 39 (1979), 3997A (Brigham Young University). Anderson, Paul V. Research into the Amount, Importance, and Kinds of Writing Performed on the Job by Graduates of Seven Univer sity Departments That Send Students to Technical Writing Courses. Paper Presented at the Annual Meeting of the Modern Language Association. Houston, December 1980. Andrews, J. Douglas. "Audience Analysis: A New Application of Decision Style Theory: A Case Study in Accounting Communica tion." A Paper Presented at the American Business Communica tion Association Convention. Phoenix, October 1981. Arapoff, Nancy. "The Semantic Role of Sentence Connectors in Extra- Sentence Logical Relationships." TESOL Quarterly, 2 (1968), 243-253. Atlas, Marshall A. Addressing an Audience: A Study of Expert- Novice Differences in Writing. Technical Report No. 3. Pittsburgh, Pennsylvania: Carnegie-Mellon University, 1979. Austin, J. L. How to Do Things With Words. Ed. F. 0. Urmson and Marina Sbisa. 2nd. ed. Cambridge, Mass.: Harvard University Press, 1975. Battison, Robbin, and Dixie Goswami. "Clear Writing Today." The Journal of Business Communication, 18, No. 4.(1981), 5-16. Benbasat, Izak, and Albert S. Dexter. Individual Differences in the Use of Decision Support Aids. Management Information Systems Research Center Working Paper Series. Minnesota: Management Information Systems Research Center, 1980. Bitzer, Lloyd T. "The Rhetorical Situation." Philosophy and Rhetoric, 1 (1968), 1-14. Bolinger, Dwight. "Accent is Predictable (If You're a Mind- Reader)." Language, 48 (1972), 633-644. 162 Britton, James. "The Composing Processes and the Functions of Writing." In Research on Composing: Points of Departure. Ed. Charles R. Cooper and Lee Odell. Urbana, Illinois: NCTE, 1978, 13-28. ___________, T. Burgess, N. Martin, A. McLeod, and H. Rosen. The Development of Writing Abilities (11-18). London: Macmillan Education, 1975. Brostoff, Anita. "Coherence: 'Next to' Is Not 'Connected to.'" College Composition and Communication, 32 (1981), 279-293. Brown, Penelope, and S. Levinson. "Universals in Language Usage: Politeness Phenomena." In Questions and Politeness: Strate gies in Social Interaction. Ed. Esther Goody. Cambridge: Cambridge University Press, 1978, 56-324. Brown, Roger, and A. Gilman. "The Pronouns of Power and Solidar ity." In Style in Language. Ed. T. A. Sebeok. Cambridge, Mass.: M.I.T. Press, 1960, 253-276. Burke, Kenneth. Counter-Statement. 2nd ed. Berkeley: University of California Press, 1968. ___________ . "introduction: The Five Key Terms of Dramatism." In A Grammar of Motives. Berkeley, California: University of California Press, 1945, iv-xxiii. Chafe, Wallace. "Givenness, Contrastiveness, Definiteness, Sub jects, Topics, and Point of View." In Subject and Topic. Ed. Charles Li. New York: Academic Press, 1976, 25-55. Charrow, Veda. Let The Reader Beware. Washington, D. C.: American Institutes for Research, 1979. . What is "Plain English," Anyway? Washington, D. C.: American Institutes for Research, 1979. Christensen, Francis. Notes Toward a New Rhetoric: Six Essays for Teachers . New York: Harper and Row, 1967. Clark, Herbert H., and Eve V. Clark. Psychology and Language: An Introduction to Psycholinguistics. New York: Harcourt Brace Jovanovich, Inc., 1977. ___________ , and S. E. Haviland. "Comprehension and the Given-New Contract." In Discourse Production and Comprehension. Ed. R. 0. Freedle. Norwood, New Jersey: Ablex Publishing, 1977, 1-40. 163 Collins, James L. , and Michael M. Williamson. "Spoken Language and Semantic Abbreviation." Research in the Teaching of English, 15 (1981), 23-35. Cook-Gumperz, Jenny, and John J. Gumperz. "From Oral to Written Culture: The Transition to Literacy." In Variation in Writ ing. Ed. M. F. Whiteman. Hillsdale, N.J.: Erlbaum, 1981. Cooper, Marilyn. "Context as Vehicle: Implicatures in Writing." In What Writers Know: The Language, Process, and Structure of Written Discourse. Ed. Martin Nystrand. New York: Academic Press, 1982, 105-128. Coulthard, Malcolm. An Introduction to Discourse Analysis. London: Longman Group Limited, 1977. Crothers, Edward J. Paragraph Structure Inference. New Jersey: Ablex Publishing Corporation, 1979. Crowhurst, Maxine. "The Effect of Audience and Mode of Discourse on the Syntactic Complexity of the Writing of Sixth and Tenth Graders." DA, 38 (1978), 7300A-7301A (University of Minne sota) . Crowhurst, Maxine, and Gene L. Piche. "Audience and Mode of Dis course Effect on Syntactic Complexity at Two Grade Levels." Research in the Teaching of English, 13 (1979), 101-109. Crystal, David, and Derek Davy. Investigating English Style. London: Longman, 1969. D 1Angelo, Frank J. A Conceptual Theory of Rhetoric. Cambridge, Mass.: Winthrop Publishers, Inc., 1975. de Beaugrande, Robert. "Psychology and Composition." College Composition and Communication, 30 (1979), 50-57. ___________. Text, Discourse, and Process: Volume IV Advances in Discourse Processes. New Jersey: Ablex Publishing Company, 1980. van Dijk, Teun. Text and Context. London: Longman, 1977. Driver, Michael J. Administrative Problem Solving Exercise (IST- RII) , 1971. ___________. "individual Decision Making and Creativity." In Organizational Behavior. Ed. S. Kerr. Columbus, Ohio: Grid Publishers, 1979, 4:1-4:33. 164 ___________, and T. Mock. -'information Processing, Decision Style Theory and Accounting Information Systems." Accounting Review, 50 (1975), 490-508. ___________, and Alan J. Rowe. "Decision-Making Styles: A New Approach to Management Decision Making." In Behavioral Prob lems in Organizations: Ed. C. Cooper. Englewood Cliffs, New Jersey: Prentice-Hall, 1980, 141-182. Ervin-Tripp, Susan, "is Sybil There? The Structure of Some Ameri can English Directives." Language in Society, 5 (1976), 25-66. Faigley, Lester, and Thomas P. Miller. "What We Learn from Writing on the Job." College English, 44 (1982), 557-569. ___________ , Thomas P. Miller, Paul R. Meyer, and Stephen Witte. Writing After College: A Stratified Survey of the Writing of College Trained People. Writing Program Assessment Technical Report Number 1. Washington, D.C.: Fund for the Improvement of Post Secondary Education, 1981. ERIC ED 210 708. Felker, Daniel B., ed. Document Design: A Review of the Relevant Research. Washington, D.C.: American Institutes for Research, 1980. : _______, Francis Pickering, Veda Charrow, V. Melissa Holland, and Jariice Redish. Guidelines for Document Designers. Washington, D.C.: American Institutes for Research, 1981. Ferguson, C. A. "Diglossia." Word, 15 (1959), 325-340. ___________. "The Structure and Use of Politeness Formulas." Language in Society, 5 (1976), 137-152. Fielden, John S. "What Do You Mean You Don't Like My Style?" Harvard Business Review, 60 (1982), 128-138. Fillmore, Charles J. "The Case for Case." In Universals in Lin guistic Theory. Ed. Emmon Bach and Robert T. Harms. New York: Holt, Rinehart and Winston, 1968, 1-88. Fish, Stanley E. "How to Do Things with Austin and Searle: Speech Act Theory and Literary Criticism." Modern Language Notes, 91 (1976), 983-1025. Flesch, Rudolf. The Art of Readable Writing. New York: Harper and Brothers Publishers, 1949. 165 Flower, Linda, and John R. Hayes. "A Cognitive Process Theory of Writing." College Composition and Communication, 32 (1981), 365-387. ______ , John R. Hayes, and Heidi Swarts. Revising Functional Documents: The Scenario Principle. Technical Report No. 10. Pittsburgh, Pennsylvania: Carnegie-Mellon University, 1980. Garnica, Olga K. "Rules of Verbal Interaction and Literary Analy sis." Poetics, 6 (1977), 155-168. Gibson, Eleanor J. , and Harry Levin. The Psychology of Reading. Cambridge, Mass.: TheM.I.T. Press, 1975. Giglioli, Pier Paola. "introduction." Language and Social Context. Ed. Pier Paola Giglioli. London: Penguin Education, 1972, 7-17. Goodin, George, and Kyle Perkins. "Discourse Analysis and the Art of Coherence." College English, 44 (1982), 57-63. Goody, Esther N. "Towards a Theory of Questions." In Questions and Politeness: Strategies in Social Interactions. Ed. Esther N. Goody. Cambridge: Cambridge University Press, 1978, 17-25. Goswami, Dixie. "Naturalistic Studies of Nonacademic Writing." In Moving Between Practice and Research in Writing: Proceedings of the NIE-FIPSE Grantee Workshop. Eds. Ann Humes, Bruce Cronnell, Joseph Lawlor, Larry Gentry. Los Alamitos, Califor nia: SWRL Educational Research and Development, 1981, 143-147. Grice, H. P. "Logic and Conversation." In Syntax and Semantics, Vol. 3: Speech Acts. Eds. P. Cole and J. L. Morgan. New York: Academic Press, 1975, 41-58. Grimes, Joseph E. The Thread of Discourse. The Hague: Mouton Publishers, 1975. Gutwinski, Waldemar. Cohesion in Literary Texts: A Study of Some Grammatical and Lexical Features of English Discourse. The Hague: Mouton, 1974. Hagen, Lyman B. "An Analysis of Transitional Devices in Student Writing." Research in the Teaching of English, 5 (1971), 190-201. Hake, Rosemary, and Joseph M. Williams. "Style and Its Consequen ces: Do As I Do, Not As I Say." College English, 43 (1981), 433-451. 166 Halliday, M. A. K. "The Linguistic Study of Literary Texts." In Essays on the Language of Literature. Eds. Seymour Chatman and Samuel Levin. Boston: Houghton Mifflin, 1967, 217-223. ___________. "Notes on Transitivity and Theme in English: Parts 1 and 2." Journal of Linguistics, 3 (1967), 37-81, 199-244. ___________. "Notes on Transitivity and Theme in English: Part 3." Journal of Linguistics. 4 (1968), 179-216. ___________, and Ruqaiya Hasan. Cohesion in English. London: Longman,.1976. ___________, Angus McIntosh, and Peter Strevens. The Linguistic Sciences and Language Teaching. London: Longmans, 1964. Hamilton, David. "interdisciplinary Writing." College English, 41 (1980), 780-796. Hellreigel, Don, and John W. Slocum, Jr. "Managerial Problem- Solving Styles." Business Horizons (1975), 29-37. Henderson, John C., and Paul C. Nutt. "The Influence of Decision Style on Decision Making Behavior." Management Science, 26 (1980), 371-386. Herrington, Anne J. "Writing to Learn: Writing Across the Disci plines." College English, 43 (1981), 379-387. Hiatt, Mary P. The Way Women Write. New York: Teachers College Press, 1977. Hirsch, E. D. , Jr. The Philosophy of Composition. Chicago: University of Chicago Press, 1977. Hymes, Dell. "Toward Ethnographies of Communication: The Analysis of Communicative Events." In Language and Social Contexts. Ed. Pier Paola Giglioli. London: Penguin Education, 1972, 21-44. Johns, Ann M. "A Comparison of Cohesive Elements In American Business and Non-Native Speaker Written Discourse." Diss. University of Southern California 1979. Kaufman, Susan D. "Cognitive Style and Writing: An Inquiry." DA, 42 (1981), 1519A (Rutgers University). Kinneavy, James L. A Theory of Discourse. Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1971. 167 Kihtsch, Walter. The Representation of Meaning in Memory. Hills dale: Laurence Erlbaum Associates, 1974. Knoblauch, C. H. "intentionality in the Writing Process: A Case Study." College Composition and Communication, 31 (1980), 160-168. . Labov, William. "The Logic of Nonstandard English." Georgetown Monographs on Language and Linguistics, 22 (1969), 1-22, 26-31. Excerpts reprinted in Language and Social Contexts. Lakoff, George. "Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts." In Papers from the Eighth Regional Meeting of the Chicago Linguistic Society. Chicago, Illinois: Uni versity of Chicago, 1972, 183-228. Lakoff, Robin. "Language in Context." Language, 48, No. 4 (1972), 907-927. ___________ . "The Logic of Politeness; or Minding Your P's and Q's." In Papers from the Ninth Regional Meeting of the Chicago Linguistic Society. Chicago, Illinois: University of Chicago, 1973, 292-305. ___________ . "The Pragmatics of Modality." In Papers from the Eighth Regional Meeting of the Chicago Linguistic Society. Chicago, Illinois: University of Chicago, 1972, 229-246. ___________ . "Tense and Its Relation to Participants." Language, 46 (1970), 838-849. Lloyd-Jones, Richard. "Primary Trait Scoring." In Evaluating Writing: Describing, Measuring and Judging. Eds. Charles R. Cooper and Lee Odell. Urbana, Illinois: NCTE, 1977, 33-66. Mann, Richard. Personal Interview. 8 June 1982. ___________ . "Unstructured Decision Making in Strategic Planning." Diss. University of Southern California 1982. McRorey, Tom J. "The Role of Personality Characteristics in Written Business Communication: A Study of the Personality Character istics of Public Relations Letter Writers in the Finance Industry in Tulsa, Oklahoma." DA, 38 (1977), 1833A (The University of Oklahoma). Mellinkoff, David. The Language of the Law. Boston: Little, Brown and Co., 1963. 168 Milic, Louis T. "Against the Typology of Styles." In Essays on the Language of Literature. Eds. Seymour Chatman and Samuel R. Levin. Boston: Houghton Mifflin Company, 1967, 442-450. ___________ . A Quantitative Approach to the Style of Jonathan Swift. The Hague: Mouton & Co., 1967. ___________ . Stylists on Style. New York: Charles Scribner's Sons, 1969. ___________ . "Theories of Style and Their Implications for the Teaching of Composition." College Composition and Communica tion, 16 (1965), 66-69, 126. Mitchell, Ruth. "Shared Responsibility: Teaching Technical Writing in the University." College English, 43 (1981), 543-555. Myers, Mildred S. "Written Communication at the Managerial and Professional/Technical Levels: A Case Study." DA, 42 (1981), 775A-776A (Carnegie-Mellon University). Nystrand, Martin. "Rhetoric's 'Audience' and Linguistics' 'Speech Community': Implications for Understanding Writing, Reading and Text." In What Writers Know: The Language, Process, and Structure of Written Discourse. Ed. Martin Nystrand. New York: Academic Press, 1982, 1-28. Odell, Lee. "Defining and Assessing Competence in Writing." The Nature and Measurement of Competency in English. Ed. Charles R. Cooper. Urbana, Illinois: NCTE, 1981, 95-138. ___________ . "What Can Evaluators Learn from Writers in Non-Academic Settings?" In Proceedings of Maryland Composition Conference, 16 April 1982. Eds. Susan Kleimann and Anne Franzak. ___________ , and Charles Cooper. "Procedures for Evaluating Writing: Assumptions and Needed Research." College English, 42 (1980), 35-43. ___________ , and Dixie Goswami. "Writing in a Non-Academic Setting." Research in the Teaching of English. 16 (1982), 201-223. ____________ , and Dixie Goswami. Writing in Non-Academic Settings. Washington, D.C.: Fund for the Improvement of Post-Secondary Education, 1981. ERIC ED 214 163. Palmer, F. R. Modality and the English Modals. London: Longman, 1979. 169 Pickering, Wilbur N. "A Framework for Discourse Analysis." DA, 39 (1979), 4219A (University of Toronto). Pratt, Mary Louise. Toward a Speech Act Theory of Literary Dis course . Bloomington: Indiana University Press, 1977. Prince, Ellen F., Joel Trader, and Charles Bosk. "On Hedging in Physician-Physician Discourse." (Unpublished, 1980). Pytlik, Betty P. "Audience: A Bibliographical Essay." A Paper Presented at the American Business Communication Association Convention. New Orleans, October 1982. Redish, Janice C., and Kathryn Racette. Teaching College Students How to Write--Training Opportunities for Document Designers. New York: Siegel & Gale, 1979. Riffaterre, Michael. "Criteria for Style Analysis." In Essays on the Language of Literature. Eds. Seymour Chatman and Samuel R. Levin. Boston: Houghton Mifflin Company, 1967, 412-430. Robinson, W. P., and Susan J. Rackstraw. A Question of Answers, Vol. 1. London: Routledge & Kegan Paul, 1972. Rose, Andrew M. , and Louis A. Cox, Jr. Following Instructions. Technical Report No. 4. Washington, D.C.: American Institutes for Research, 1980. Rowe, Alan J. "Decision Making in the '80*s." Los Angeles Business and Economics (1980), 7-9. ___________. Decision Style Inventory III. 1981. ___________, Richard 0. Mason, and Karl E. Dickel. Strategic Man agement and Planning: A Methodological Approach. Boston, Mass.: Addison Wesley, 1982. Rubin, Donald L. "The Development in Syntactic Aspects of Audience Adaptation Skills in Written Persuasive Discourse." DA, 39 (1979), 7201A (University of Minnesota). ___________, and Piche, Gene L. "Development in Syntactic and Strategic Aspects of Audience Adaptation Skills in Written Persuasive Communication." Research in the Teaching of En glish, 13 (1979), 293-316. Sacco, Jodee. The Language Simplification Movement: Where Does it Stand? 1980. ERIC ED 193-634. 170 Schwartz, Helen J. Teaching Stylistic Simplicity with a Computer ized Readability Formula. Paper Presented at the American Business Communication Association Convention. Washington, D.C., December 1980. Scott, William A. "Reliability of Content Analysis: The Case of Nominal Scale Coding." The Public Opinion Quarterly, (1955), 321-325. Searle, John R. "A Classification of Illocutionary Acts." Language in Society, 5 (1976), 1-23. ___________ . Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press, 1969. ___________ . "What Is a Speech Act?" In Philosophy in America. Ed. M. Black. Ithaca, New York: Cornell University Press, 1965, 221-239. Sinclair, J. M. , and R. M. Coulthard. Towards an Analysis of Discourse: The English Used by Teachers. London: Oxford University Press, 1975. Smeltzer, Larry R. "The Relationship Between Writing Style and Leadership Style." The Journal of Business Communication, 18 (1981), 23-32. ___________ . "A Study of the Relationship Between Leadership Style and Writing Style for Supervisors and the Effects of Super visory Experience, Formal Education and a Management Training Program on This Relationship." DA, 41 (1981), 2898A (Northern Illinois University). Soldow, Gary F. "A Study of the Linguistic Dimension of Information Processing as a Function of Cognitive Complexity." The Journal of Business Communication 19, No. 1 (1982), 55-69. Steinmann, Martin, Jr. "Speech-Act Theory and Writing." In What Writers Know: The Language, Process, and Structure of Written Discourse. Ed. Martin Nystrand. New York: Academic Press, 1982, 291-323. Stine, Donna, and Donald Skarzenski. "Priorities for the Business Communication Classroom: A Survey of Business and Academe." Journal of Business Communication, 16 (1979), 15-30. Stubbs, Michael. "Written Language and Society: Some Particular Cases and General Observations." In What Writers Know: The Language, Process, and Structure of Written Discourse. Ed. Martin Nystrand. New York: Academic Press, 1982, 31-55. 171 Thorne, Barrie, and Nancy Henley, Ed. Language and Sex: Difference and Dominance. Rowley, Mass.: Newbury House Publishers, 1975. Wells, Rulon. "Nominal and Verbal Styles." In Style in Language. Ed. T. A. Sebeok. Cambridge, Mass.: M.I.T. Press, 1960, 213-220. Widdowson, H. G. Teaching Language as Communication. Oxford: Oxford University Press, 1978. Williams, Joseph M. "Non-Linguistic Linguistics and the Teaching of Style." In Linguistics, Stylistics, and The Teaching of Composition. Ed. Donald.McQuade. Akron, Ohio: University of Akron, 24-40. ___________. Style: Ten Lessons in Clarity and Grace. Glenview, Illinois: Scott, Foresman, 1981. Winterowd, W. Ross. "The Grammar of Coherence." In Contemporary Rhetoric: A Conceptual Background With Readings. New York: Harcourt Brace Jovanovich, Inc., 1975, 225-233. Witte, Stephen, John A. Daly, Lester Faigley, and William R. Koch. The Empirical Development of an Instrument for Reporting Course and Teacher Effectiveness in College Writing Classes. Writing Program Assessment Project Technical Report Number 3. Wash ington, D.C.: Fund for the Improvement of Post-Secondary Education, 1981. ERIC ED 211 981. ' _______ , and Lester Faigley. "Coherence, Cohesion, and Writing Quality." College Composition and Communication, 32 (1981), 189-294. Zmud, Robert W. "On the Validity of the Analytic-Heuristic Instru ment Utilized in 'The Minnesota Experiments'." Management Science, 24 (1978), 1088-1092.

Linked assets