The conventional wisdom about buffers and negative messages needs rethink-ing. Textbook advice is often simplistic and frequently wrong, fostering inef-fective formulas and ritualistic responses. Recent research on cognitive com-plexity and social perspective-taking suggests the need for more sophisticatedaudience analysis protocols for dealing with the negative message genre.

AN INTERPERSONAL APPROACH TOWRITING NEGATIVE MESSAGES

Douglas SalernoWashtenaw Community College

WRITING NEGATIVE MESSAGES IS ONE of the most difficulttasks facing business communicators. Because we usually find say-ing "no" harder than saying "yes," and because refusing a requestoften is interpreted by a reader as personal rejection, most writersknow enough to approach the task of writing negative messageswith some degree of caution. Recently I spent all of five minuteswriting a note to a job applicant, telling her that she had beenhired. I spent almost ten times as long composing the letter sent tothe applicants who did not receive the job offer.

WHY A BUFFER OPENING?

Most business communication textbooks talk about strategiesfor writing negative messages, including refusal letters. In their dis-cussion, most of these textbooks offer tips on writing the introduc-tory paragraph or buffer. The buffer, as most textbooks present it,serves to prepare the reader for the negative news. It is almost al-ways the first pEiragraph and contains the information which either"softens" the reader for the bad news to follow or attempts todiminish the negativism of the rest of the letter.

In this essay, I want to examine some commonly held (andtaught) notions about buffer writing and to suggest how we mightbetter inform ourselves on this topic by considering research in au-dience analysis and interpersonal communication. My main point istwofold: not all negative messages require a buffer; and, in teachingbuffer writing to our students, we should be more concerned withwhy they construct them than with how they construct them.

41

42 THE JOURNAL OF BUSINESS COMMUNICATION * 25:1 :WINTER 1988

CONCEPTS OF AUDIENCE

Instruction in writing buffers assumes a concern for audience andaudience analysis. This concern, in turn, presupposes that writerscan shape their language in specific ways to reach their readers and,most importantly, that all readers are not the same. Yet when text-book writers talk about audience analysis they often assume thattheir student readers already have mastered this difficult skill. MyraKogen, for example, notes that most textbooks treat the writer'srelationship to audience "in peripheral introductory chapters thatdescribe audience contact in terms of send-message-receiver modelslargely borrowed from oral communication."' She elaborates two"paths" that current reseairch on audience is taking. The first pathconceptualized audience as a "real" person or group of people. We"analyze" this audience by gathering information about them. Sucha viewpoint encourages us to "identify the kinds of audiences forwhich a work is intended so that discourse can be shaped to theneeds of particular readers" (p. 2). Thus we try to shape the mes-sage to take into account the needs, attitudes, and other pertinentbackground information of that specific group of readers. We prob-ably will construct a different message for a group of conservative,middle-aged women than for an audience of liberal young women;for example, messages to friendly audiences will differ from mes-sages to hostile or indifferent audiences.

An alternate conceptualization treats audience as a "fiction."Walter Ong^ typifies this research path. Because we can never"know" our audience completely, we "create" those aspects thatare missing from our knowledge of our reader. We encourage ourreader to "assume a stance" with us. Assuming the dual roles ofplayviT^ht and principal actor, we feed our audience "cues" as tohow we expect them to react to our "lines." For example, in writ-ing a letter attempting to regain the goodwill of a present customerwho has suffered poor service from our organization, we may "cre-ate" a reader who is good natured, forgiving and willing to give usanother change. Thus, rather than "bombarding" our reader withlogic and good reasons as to why they should "give us anotherchance," we create a situation with our fictive audience which al-lows them to persuade themselves.

Kogen's typology parallels the recent work of Lisa Ede andAndrea Lunsford^ who develop the notion of "audience addressed"and "audience invoked." "Those who envision audience as ad-dressed," they argue, "emphasize the concrete reality of the writer's

INTERPERSONAL APPROACH TO NEGATIVE MESSAGES 43

audience; they also share the assumption that knowledge of thisaudience's attitudes, beliefs, and expectations is not only possible(via observation and analysis) but essential" (p. 156). On the otherhand, writers who invoke an audience do not "deny the physicalreality of readers" but simply assume that they "cannot know thisreality in the way that speakers can." Thus in adapting a message toan audience, the writer "uses the semantic and syntactic resourcesof language to provide cues for the reader—cues which help to de-fine the role or roles the writer wishes the reader to adopt in re-sponding to the text" (ibid). Unlike Kogen, Ede and Lunsford donot view these two audience types as contradictory or dichoto-mous. In their attempt to reconcile what they see as an inherentweakness in adapting one perspective to the exclusion of the other,Ede and Lunsford propose a model which embraces both "paths"or audience viewpoints.

Finally, Barry M. Kroll proposes three perspectives on audience."The first perspective, the rhetorical, is similar to "audience ad-dressed." Kroll notes that in this most common approach to audi-ence analysis, textbooks often "provide lists of general humancharacteristics along with questions designed [to] help the novicewriter decide which of the characteristics—intelligence, socio-economic status, occupation, educational level, and so forth—aremost relevant for reaching a particular audience in a particularrhetorical situation" (p. 173). Second, the informational perspec-tive, treats audiences as information processors. "In simple terms,"KroU says, "the writer aims to get information into the reader'shead" (p. 176). From this perspective, "readers 'process' messages,transforming linguistic input into a conceptual code that must beintegrated with information already stored in memory. Since thegoal of writing is to get information into that memory store, awriter needs to understand how the reading process works, payingparticular attention to the kinds of difficulties readers encounter intheir efforts to extract information from texts" (ibid). Third, thesocial perspective, treats writing as a "social activity"; thus whatwriters say, and how they say it, is shaped in part by the culture inwh zh writers live. In the classroom setting, this perspective isimplemented in part by having students read their writing aloud tofellow classmates. One key element in this viewpoint is "decenter-ing" or "the ability to escape from a focus on one's own perspective,especially to avoid the 'egocentric' tendency to impute this perspec-tive to others" (p. 179).

Little of the preceding research is addressed in current textbooks.

44 THE JOURNAL OF BUSINESS COMMUNICATION * 25:1:WINTER 1988

which assume that students need only consider certain demographicinformation and they automatically will be able to address an effec-tive piece of communication to their audience. Thus student writ-ers, and even their counterparts in the business world, often use thebuffer in a manner which is mechanical and obvious, sometimeseven exacerbating the very problem they are trying to solve.

CONVENTIONAL WISDOM

Much writing on the buffer is ineffective for two reasons. First,some of the information is just plain wrong. Textbook writers usu-ally assume that all bad-news communication requires some kind ofbuffer and/or indirect way of presenting the negative message. Forexample, one textbook warns: "Readers never want to hear 'no' ina letter, for they desire and expect a 'yes' answer to a request. Forthis reason, never begin a 'no' letter with the refusal."' Anotherbook, however, advises: "not all refusals are likely to upset thereader. Sometimes requests reflect only casual interest; other timespeople ask for something simply to see whether or not it's available. . .. Quite frankly, using the bad-news strategy isn't always worththe effort, even when you know the message will arouse feelings ofdisappointment."^ A second problem occurs when textbookspresent the buffer as a manipulative vehicle, suggesting that writers"trick" their audience into reading something that isn't there. Forexample, one text says "The fact is that you start bad-news mes-sages by 'beating around the bush,' but you do it skillfully. Beatingaround the bush has a bad reputation because most people do it soclumsily that it's quite obvious. It works only when the readerdoesn't notice what you're up to. By the end of this chapter, you'llbe an expert at it."^

Another textbook, analyzing an example of a buffer in a bad-news letter, offers the following comments: "The thank yousounds sincere, is germane to the invitation, and makes a positivestatement about volunteerism" [emphasis mine] .* Such a viewinvites student reader cynicism. How else can we reconcile commen-tary which suggests that the "skill" of buffer writing is not in whatwe do but in how well we conceal what we are trying to do? "Beat-ing around the bush" has a bad reputation not because "mostpeople do it so clumsily" but because most people view it as manip-ulative and unethical.

Presentations of the buffer often depict the writing situation as

INTERPERSONAL APPROACH TO NEGATIVE MESSAGES 45

an adversarieil one between writer and reader. Students are advisedto "be positive" or "don't be so bluntly negative" in constructingthe opening paragraph. Given the tenor of this writing problem, it isno wonder that students develop a cynical view of the entire com-munication process. Such teaching merely affirms what many stu-dents, and other practitioners, already believe: that good, effectivebusiness writing is manipulative, that it isn't what you say—andespecially it isn't why you say it—but most importantly how yousay it that matters. Style over substance. Besides breeding cynicism,such a teaching approach at best reinforces a wrong view of com-munication (mechanistic, manipulative) and at worst trivializes thecomplicated cognitive, social, and ethical decisions that go into theformation of a piece of writing.

Let's examine one kind of "bad news" message with which al-most all of us are familiar. Many of us, myself included, have beenboth on the receiving and sending end of these messages: the jobrefusal letter. To begin, we need to consider two important points:the context in which the letter is written and the audience's per-spective in reading the letter. Most experienced job seekers soonrealize that the arrival of a written response to a job application let-ter almost certainly means bad news. Employers typically use thetelephone to communicate job offers and employment interviews;they use the mail to send rejection notices. In most cases, then, theletter reader is expecting "negative news" even before opening theenvelope. Thus a "beating around the bush" buffer may intensifythe ill will we are trying to avoid in the reader. The sheer number ofrefusal letters that our reader may have received makes our job allthe more difficult. The potency of a stock buffer is severely dimin-ished after the reader has received a half dozen or so refusal letters.The reader soon comes to appreciate, even anticipate, a certainstructural technique: buffer, rejection, reason, goodwill closing. Formany readers, then, the refusal letter can lose most or all of itsfunctional communicative effects and become instead merely aritualistic device, similar to typical greetings between co-workers atthe office.

ANALYSIS OF JOB-REFUSAL LETTERS

To illustrate my point, I would like to share the results of ananalysis of twenty-two rejection letters I received in the last sixmonths of 1983. The letters came both from academia and private

46 THE JOURNAL OF BUSINESS COMMUNICATION ' 25:1 :WINTER 1988

industry and were in response to a variety of jobs advertised forwhich I sent a job application letter. Eight of the letters containedno buffer at all. In two cases, the writer emphasized the bad newsby beginning with "I am sorry to inform you . . ." or "It is my un-pleasant duty . . .." The other six letters began without any socialamenity, merely stating the result of their job search. Four such let-ters began: "We have now come to the point . . ."; "I write to tellyou that . . ."; "We have received your inquiry . . ."; and "We haveat last finished . . .." Eight other letters contained a buffer butpresented the bad news closely afterwards, diminishing the buffer'seffect. Typically the letter began with "Thank you" and then in thesame sentence or in the very next sentence used language such as"however," "but," "unfortunately," "I regret," "we regret, how-ever," and "I am sorry to tell you." In these cases, the letter writerseemed to regard the buffer as a mere formality, as ritualistic lan-guage. Readers would be likely to miss the intended buffer com-pletely because of the proximity of the negative news.

Finally, six letters contained at least a full-sentence buffer and a"bridge" between the opening paragraph and the bad news. Thewriter usually used several paragraphs to explain the selection pro-cess. In some cases, the name ofthe appointed person was given. Inall cases, the writer closed by repeating some part of the buffer,usually the "thank you", and wished the reader well ("good luck")in the job search. Two letters stand out as being particularly wellcrafted. The first message was typed on a postc£ird and consisted ofonly two sentences: "We have at last finished our review of applica-tions and have filled the position in our department. Thank you forapplying." The second letter began "We are most happy to an-nounce that [name of appointee] has accepted a position . . .." Thefirst letter's effectiveness is contained in its brevity. The letter writ-er, perhaps realizing the ineffectiveness of a letter easily recogniz-able as belonging to the "refusal genre," finesses the expected re-fusal by allowing the reader to infer that "I did not get the job."Such a tactic removes the necessity of the almost always awkward(and insincere) language extolling the reader's virtues and merits.The tone and brevity confirm a reasonably friendly, humane refusal.The second letter also catches the jaded refusal letter reader by sur-prise. As in the first case, the writer spares the reader a direct refus-al and instead allows the reader to infer the bad news through theinformation provided by the writer.

Given the above, we might want to rethink the notion that allbad news messages necessarily require a buffer. Particularly when

INTERPERSONAL APPROACH TO NEGATIVE MESSAGES 47

our reader expects to receive bad news, the reader may interpretthe buffer as either insincere or merely ritualistic, neither responsebeing what we want. Certainly there are other kinds of businesscommunication where the telephone is used to communicate com-pliance and where a letter conveys a refusal. In these cases, the buf-fer may be irrelevant, even counterproductive in maintaining goodwill.

There are many situations, however, when a reader opens a let-ter and either does not know what to expect or has optimistic ex-pectations. In these situations, a well-written buffer can prepare thereader for refusal. However, the writer will need to be sensitive tothe reader's expectations and motivation for making the requestand the likely response to the refusal (anger, disappointment, resig-nation) and the reader's "next step" as a consequence of receivingthe refusal (litigation, another letter, withdrawal of business). Inpreparing a written piece of communication, we behave in wayssimilar to facing an interpersonal exchange, viewing the letter notfrom a one-way perspective, but anticipating the interior state ofour audience and their likely response to certain strategies.

AN INTERPERSONAL PERSPECTIVE

I would like, then, to present an interpersonal approach to writ-ing negative messages. My perspective presupposes that writing in-struction and evaluation should concem itself not only with theproduct, the what that students produce, but also the process,trying to discover why students produce the writing they do. ForexEimple did they complete a business letter assignment merely bycopying some format outlined on the blackboard, did they copy itfrom 1001 Models of Effective Business Communication or didthey actually sit down, analyze the situation and audience, and thenattempt a specific response to it. My instructional perspective isbased upon recent research and writing on cognitive complexity,perspective-taking, and audience.

Cognitive complexity research generally is recognized as begin-ning with the work of psychologist George Kelly' who argued thatpeople's processes are "psychologically channelized by the ways inwhich [they] anticipate events."*" As Michael Beatty and StevenPayne have noted, most cognitive complexity research rests on thenotion that individuals differ in the number and kinds of "personal

48 THE JOURNAL OF BUSINESS COMMUNICATION • 25:1 :WINTER 1988

constructs" or "lenses" they use to view and anticipate events andprocess information." Jesse Delia and Barbara O'Keefe define con-structs as "organized schemas or patterns of expectation withinwhich events are construed or interpreted." "Any event," they ar-gue, "can make sense only in so far as it is ordered within the con-struct system."'^ Researchers typically measure cognitive complex-ity by having subjects write descriptions of two persons they know,one they like and one they dislike. Researchers then count the num-ber of different ways subjects describe the two people. Cognitivecomplexity can be viewed as a continuum of information-processingability, specifically information about people. According to re-search summarized by Beatty and Payne, the lower end of the con-tinuum includes "categorical thinking, rapid closure, and simple andabsolute rules for combining constructs whereas abstract, contin-uous, probabilistic thinking, complex and flexible combinatoryrules, and suspension of closure are indicative of the higher end ofthe continuum.'"' Researchers argue that cognitive complexitymeasures are different from measures of intelligence or verbalacuity, and that this skill is developmental, increasing with chrono-logical age.

Perspective taking describes "the ability of a person to representanother's perspective or point of view.'* Research with children £indadolescents conducted by Delia, Clark, and others suggests at leastthree levels of perspective taking. At the lowest level, subjectsgive no evidence of perspective taking at all. Persuasive strategiesconstructed at this level are entirely centered within the perspectiveof the persuader. At level two, subjects acknowledge that their audi-ence does not share their opinion, but subjects provide argumentswhich support only their own viewpoint. At level three, subjectsnot only anticipate objections but also provide counter-Eirgumentsin messages to their audience.'^ While statistical analyses of cogni-tive complexity and social perspective taking results have not pro-vided researchers with a unanimous conclusion regarding thecorrelation between these two measures, some writers suggest sucha relationship." Finally, severtd studies conclude that there is a pos-itive correlation between cognitive complexity and communicationeffectiveness.*'' As O'Keefe and Sypher argue in their critical re-view of cognitive complexity research, this measure has "importantconsequences" for the kinds of communicative strategies, rationalesand number of persuasive arguments used by individuals in persua-sive encounters with their audiences.'^

INTERPERSONAL APPROACH TO NEGATIVE MESSAGES 49

RESEARCH-BASED STRATEGIES

The research reviewed above suggests three strategies to considerwhen we write negative messages. First, we should realize that notall audiences are the same. If the purpose of our communication isto refuse a request and to retain our reader's goodwill, then wemust obtain a sense of that person as a unique individual. It followsthen that we should try to assume the perspective of the person towhom we are communicating. This involves more than merely pre-senting socifd niceties in the introduction of the letter; instead, weneed to provide sufficient information and cues which allow ourreader to feel understood and addressed personally. Finally, wemight benefit from invoking an audience rather than trying to"manipulate" some preconceived audience. Our writing would thenprovide "cues" to our readers, informing them of a perspective,establishing a stance from which we and they could complete themost effective and efficient communication transaction.

The teaching strategy I use to present effective buffer writing(and all other writing) is based on Roundy and Thralls' work, whichallows students to construct their own writing assignments." Mystudents construct a plausible problem situation which necessitatesa letter. They specify an audience for their assignment and a speci-fic purpose they hope to achieve with that letter. When they havecompleted this preliminary assignment, they have created a "casestudy" for a business letter. I evaluate their case study, grading pri-marily on thoroughness of formulating the problem. Then, afterreceiving my comments on their case study, they write the businessletter. I evaluate that assignment on the basis of information pre-sented in their case study. My methodology assumes that no oneknows the audience better than the writer. Therefore, I can bestevaluate someone's writing by looking over their shoulder, by hav-ing the student inform me of the specifics (problem, audience, pur-pose) of the writing context. The buffer may or may not be neces-sary in a particular writing context. The student may, for example,create a reader who likes his or her messages "straight and to thepoint." I let the student decide that. If a buffer is necessary, if"space" is needed to put the reader in a mindframe conducive toaccepting a negative message, I let the student decide on the natureof the buffer. Thus a "thank you for your request" may be an ap-propriate buffer not because it "beats around the bush" but be-cause it is something the writer has decided that his or her readerexpects in a business letter, or because the reader will want some

50 THE JOURNAL OF BUSINESS COMMUNICATION ' 25:1 :WINTER 1988

acknowledgment of his or her letter before receiving a reply. ThusI am just as interested in my students' writing strategies as I am inthe content of their message.

I have tried to a i ^ e that letter writing is not simply informationtransfer. Research in interpersonal communication and audienceanalysis suggests that we examine all writing strategies taught to ourstudents. We m^ht reexamine our rationale for using a buffer, themanner in which we define its use and thus present a slightly moresophisticated perspective of the communication process. The re-search on cognitive complexity, perspective taking and audienceanalysis does not fully inform us on how to be effective communi-cat> r̂s in business letters. But this research can broaden the perspec-tive we present our students regarding the communication process.

NOTES

' Kogen, Myra (1983, December). The role of audience in business andtechnical writing. The ABCA Bulletin, 46, 2-4.

^ Ong, Walter (1975). The writer's audience is always a fiction. PMLA,90,247-257.

^ Ede, Lisa & Lur.sford, Andrea (1984, May). Audience addressed/audienceinvoked: The role of audience in composition theory and pedagogy. CollegeComposition and Communication, 30,155-171.

"* Kroll, Barry M. (1984, May). Writing for readers: Three perspectives onaudience. College Composition and Communication, 30,172-185.

^ Brown, Leland L. (1982). Communicating facts and ideas in business(3rd Ed.). Englewood Cliffs, NJ: Prentice-Hall.

* Hatch, Richard (1983). Business communication: Theory and technique.Chicago, IL: Science Research Associates.

•' Hatch, p. 247.^ Rutkoskie, Alice & Murphree, Carolyn T. (1983). Effective writing for

business. Columbus, OH: Charles E. Merrill.' Kelly, George A. (1955). The psychology of personal constructs (2 vols.).

New York, NY: Norton.' " Kelly, George A. (1963). A theory of personality: The psychology of per-

sonal constructs. New York, NY: Norton.' ' Beatty, Michael J. & Payne, Steven K. (1984, Summer). Loquacity and

quantity of constructs as predictors of social perspective-taking. Communica-tion Quarterly, 32, 207-210.

' ^ Delia, Jesse G. & O'Keefe, Barbara J. (n.d.). Constructivism and commu-nicative development. Unpublished manuscript.

' ^ Beatty & Payne, p. 209.'"* O'Keefe, Daniel J. & Sypher, Howard E. (1981, Fall). Cognitive complex-

ity measures and the relationship of cognitive complexity to communication.Human Communication Research, 8, 72-92.

INTERPERSONAL APPROACH TO NEGATIVE MESSAGES 51

' ' Delia, Jesse G., Kline, Susan L., & Burleson, Brant R. (1979, November).The development of persuasive communication strategies in kindergartenersthrough twelfth graders. Communication Monographs, 47, 241-256.

' * O'Keefe & Sypher, pp. 81-82.* •̂ O'Keefe, Barbara J. & Delia, Jesse G. (1979, November). Construct com-

prehensiveness and cognitive complexity as predictors of the number and stra-tegic adaptation of arguments and appeals in a persuasive message. Communica-tion Monographs, 47, 230-240; Daniel J. O'Keefe & Howard E. Sypher (1981,Fali); Jesse G. Delia, Susan L. Kline, & Brant R. Burleson (1979, November);Jesse G. Delia, Ruth Anne Clark, & David E. Switzer (1979, November). Thecontent of informal conversations as a function of interactants' interpersonalcognitive complexity. Communication Monographs, 47, 274-281; Claudia L.Hale (1980, November). Cognitive complexity-simplicity as a determinant ofcommunication effectiveness. Communication Monographs, 47, 304-311.

* ^ O'Keefe & Sypher, pp. 85-86.' ' Roundy, Nancy & Thralls, Charlotte (1983, Summer). Modeling the com-

munication context: A procedure for revision and evaluation in business writ-ing. The Journal of Business Communication, 20, No. 3, 27-46.