Introduction
Efforts at influencing public attitudes toward the risks associated with the HIV virus stand at the crossroads of many different discourse communities. Discourse on HIV ranges from that in highly technical scientific journals and papers delivered at scientific meetings, to newspaper coverage, to tabloid finger-pointing and the cure-of-the-week stories. Some of these pieces reach small audiences, others larger ones. Though many risk communicators work outside the AIDS education field, or even outside public health education, communications about AIDS are paradigmatic of problems encountered by anyone attempting to reach the general public in the increasingly complex scientific context of our century.
The single piece with the largest documented reach -- and that is not to enter into the vexed question of effect -- is surely the AIDS National Mailing (ANM) of 1988. In June of that year, the Congressionally-mandated ANM arrived at approximately 108 million American homes. While provisional data from the National Health Interview Survey and Gerbert and Maguire's study1 suggest that it had an impact in public health education terms, and while the Macro Systems study2 provided a formative evaluation with particular emphasis on the intermeshing of scientific, health education and advertising personnel, and Ledbetter and Johnson3 have studied its reading levels, the document itself has not received the attention it deserves in public AIDS discourse. Can one best understand the contribution of the ANM from the viewpoint of theorists of risk communication, of social marketing, or of cultural studies?4 Though this study began as an effort to find models of effective workplace composing for college students, it rapidly involved learning about scientific discourse, public policy questions, risk communication theory, and the history of sexuality, all of which figured importantly to the ANM writers who struggled to make decisions draft to draft.
Attempting to understand the significance of this text for risk communicators specifically, this study looked carefully at both the process which produced the ANM and the movement of the drafts through various revisions. What emerged as central was a movement away from impersonal treatments of the material, which distanced authorial voice from audience, and toward an increasingly personalized tone, which lessened the distance between the authorial voice and the implied audience. As previous scholars of public communications have noted,5 as the authorial voice is no longer one authoritarian voice, the audience is encouraged to see itself not simply as a depersonalized "receive site," but as human beings with diverse lives and with the potential for empowered action. Since one of the key variables in the Health Belief Model is "self-efficacy,"6 defined as the perception that it is within the person's power to do something about the risk, any effort at communicating risks which are associated with the HIV virus and changing health behaviors associated with them will, in fact, be acting according to this model in increasing the emphasis on empowering self-hood within the text. The "self," after all, must precede the operation of "self-efficacy." While the writers of the ANM knew that their first goal was general attitudinal change -- rather than specific behavioral changes by 107 million Americans -- they also realized the centrality of "self-hood" to the empathetic effort as well.7
In the depersonalized landscape of disease discourse, the ANM, both in the process of its creation and in the final product, emphasized the specific identities of various groups, whether writers speaking out of their expertise or individuals created as an implied audience. The path to empathetic unity led directly through an emphasis on diversity.
First, in the process of its composition, voices from different audiences were encouraged, and the document was cycled in such a way that a group with one expertise was always replying to a group with a contrasting expertise. The model was not a melting pot but a dialogue of diverse voices. Though the final product might be described as a one-way communication, from "experts" to a target audience, that which Ann Fisher describes as the "informing" not the "empowering end of a spectrum of relations to the audience,"8 the creation of audiences within the composing process itself meant a recognition of the existence and differences of persons.
Second, the assumptions about the target audience changed throughout the drafts. After introducing some representative concepts of audience from contemporaneous documents, this study will go on to describe the composing process and consider changes in three important sections of the ANM, from first to final drafts, and will show a move away from the authoritarian and impersonal and toward the levelling and empathetic voice.
Method
This study could not have been conducted without the assistance of the National AIDS Information and Education Program at The Centers for Disease Control and Prevention (CDC) in Atlanta and the Atlanta office of Ogilvy & Mather. These two groups provided the author with drafts, chronologies of the process, ancillary correspondence, and interviewing time on site.9 Beginning in late December 1987, drafts were circulated between writers at Ogilvy & Mather's Atlanta office and individuals at CDC, including both scientists asked to review particular concepts and others involved specifically in the National AIDS Information and Education Program. Audience reaction to the material was gathered in a number of ways.10
The goals of this study were two-fold: first, to identify some significant changes along the way to the final draft, and, second, to see if the source of those changes could be accounted for. The methods used in this study are those of a student of verbal and cultural style, not those of a social scientist.11 AIDS is, itself, no respecter of disciplinary boundaries. In offering this study to those specifically interested in risk communication, however, the author is aware of being a minority voice, one moving not from science toward communication, but from communication toward science. Rhetorical analysis, from ancient Greek rhetoricians to the present, has often considered audience central but those studies of audience have worked by inferential methods from the text itself rather than proceeding by social scientific measurement of the audience. For rhetorical analysis, the analyst works to answer the question: who wants information such as this, whose assumptions or prejudices are being considered here. It works backwards from the texts: If X appears in the text as an understandable statement, what sorts of understanding/assumptions are needed to make sense of X? Sometimes rhetorical studies may involve establishing the stylistic context of particular forms. Workplace ethnography goes to the writers and talks to them about exactly what they are thinking when these stylistic choices are made. Retrospective ethnography, however, is vexed. Once one is asking not "Why are you doing this? " but "Do you remember why you did this then?" the responses are blurred, and what we are looking at may be the construction of the writer's mind, a script about how they like to think they write. The field of workplace ethnography is growing in graduate programs teaching writing, but a study such as this one, initiated at its earliest moment more than a year later than the writing event itself, cannot pretend to ethnography except analogously.
Also, on this particular topic, at this particular time, so many factors influence the decisions writers made. AIDS is not a subject without extraordinary resonance: social, ethical, scientific, and political, just to begin the listing. Therefore, if one is not present at the moment of composition, then one's work becomes a study of how the scholar comes to understand what someone made of AIDS in writing about it. Certainly most risk communicators deal with subjects of troubling resonance. In a few instances, the risk communicator may be working with a subject where the principle trouble is that the audience does NOT have a problems with the topic, e.g., radon. Most risk communications, however, will be issued in situations where emotional and political and scientific issues are complex and clouded. So, though not every risk communicator thinks about AIDS, AIDS risk communication seems almost the paradigmatic case to consider. That the field of risk communication itself is becoming more interested in the cultural questions involved in communicating risks -- whether they be of cultures national, international, or corporate -- suggests that there is room for the study of risk communication efforts from within the humanities, as well as from within the sciences, efforts which insist, in the words of Roger Kasperson, that "the conduct of risk communication is embedded in its mission as a humane enterprise."12
The Implied Audience: Available Health Education Models
The range of implied audiences found in health education documents can be demonstrated by looking at representative documents from the period of the ANM, one from a state board of health,13 another from a national pharmaceutical organization14 and an earlier document issued by the Surgeon General.15 In looking for evidence of the implied audi-ence, whether a text chose a structure of unified or discreet elements, used logical or emotional evidence, used coherent or disjunctive sequencing and made particular dictional choices have been considered.
First, a brief pamphlet -- an 8 and 1/2 inch sheet folded into six panels -- from the State of Indiana.16 The verbal text is very brief. Red is used for the heading of the subsections and for emphatic phrases: "AIDS KILLS -- Get Tested -- AIDS KILLS." There is no punctuation at the end of statements, and the bulleted subheads do not have to be read in order. Even the ending and beginning headlines could be interchanged. Does some particular need of an audience seem to determine the style? It appears, in fact, to be not the receiving audience but the writer's ease, the ease with which the writer can list material, which accounts for the structure. For example, in the interests of parallelism and simplification, a section is headed "You can get AIDS if you..." and followed by a number of statements concluding that sentence. Consider, however, who is the implied audience when the statement is completed with "are a baby born to a woman who has the AIDS virus." Since babies don't read, the attention is directed not to the reader, but to the writer. Throughout, statements are set up in conditional or imperative structures ("if"s or "don't"s) which maintain the author as imperial and the advice as simple. Though there is an implied appeal to logic in the repeated use of "if/then" constructions, the overarching appeal is emotional: "AIDS KILLS" stands at the beginning and end. What the conditional statements in between really insist upon is that the writer controls the disease, controls the baby, and can make promises about what will happen "if." Though it is possible to imagine that the sections are read out of order, the basic left to right reading assumption seems to rule here, and, with it, a judgment on the audience's ability to pick out what it needs. The visuals are cartoon simplifications, and suggest that the only audience are those who attend wild parties. Instead of even trying to establish metaphorically that AIDS presents a bill for prior activities, the pamphlet makes a literal situation its focus, thus allowing the many readers who do not fit into the groups depicted to comfortably establish a distance from the disease. In its dictional choices, this pamphlet avoids scientific terminology or statistics, but incorporates street terms only in parentheses after the initial Latinate anatomical terms. For example, "anal (Greek)," "vaginal (straight)," "oral (French)," and "condoms (rubbers)." The structure insists upon the author's right to establish which is the primary and which the secondary term, and has the unfortunate effect of distancing readers who quibble with the parenthetical definitions. Are all vaginal sexual encounters straight? French kissing and French sex are apparently defined as the same, not to mention the agitation of the Greek community at reading "Greek." Occasional terms more recondite than "street," such as "nonoxynol-9" or "spermicide," are used, but "works" is not defined at all. The sentence "The AIDS test is simple" exemplifies the approach of the brochure. Simple, yes, insofar as there is no complicated procedure which the patient undergoes, but simple, no, if we're talking about the sequence of tests recommended or the false positive problem or the questions of confidentiality. The Indiana pamphlet simplifies AIDS, because its writers appear to have assumed superiority to their audience, and both the certainty and elevation of a thundering all-caps prophet: "AIDS KILLS."
The pharmacy pamphlet increases the sophistication of the appeal, and the implicit definition of the audience changes from thoughtless party animals to scared heterosexuals.17 The 8-sided pull-out is printed on pale taupe heavyweight bond, and the few illustrations are either silhouettes of the lateral view of a naked outline favored by height and weight charts, or pen and ink drawings of faces of worried individuals -- two males, two females, one of whom is African-American. All information is conveyed in complete sentence units, the majority being simple sentences. The complex sentences are if/when constructions, and simple sentences are often introduced by modifying phrases such as "In order to," or "By doing [x]..." There is a clear preference for logical evidence. The first page tells us "Nonsense can't stop AIDS." The pamphlet sees the audience's "problem" as not having enough information, and thus chooses questions as a useful way to proceed. The initial heading reads "AIDS: Now it's everyone's concern." While this might seem to trumpet community, the pharmacists' community, working to save "YOU" from getting this disease, is lauded. Though the pamphlet was designed for distribution in the twelve cities most affected by AIDS, the audience is clearly the well, but worried, heterosexual. "Protection" and "control" are recurrent themes. The audience is clearly assumed to have greater reading skills than in the Indiana pamphlet, and no street terms are used. Without parenthetical definition, readers encounter "intravenous equipment," "antibodies," "syphillis," "herpes," "chancroid" and "hemophiliacs." Only occasionally is parenthetical or additive definition used: "antibodies, special substances which attack foreign substances." AIDS itself is discussed as having stages (HIV positive, ARC, AIDS), and there is a brief discussion of symptoms, both in the ARC and AIDS stages.
The Surgeon General's Report on Acquired Immune Deficiency Syndrome represents yet a third style of conveying information about AIDS.18 The black and white cover of this 36-page pamphlet is adorned only with the logo of the Public Health Service and the title. Here we find for the first time many of the themes to be picked up in the ANM, especially the emphasis on the fear in the hearts of most Americans. This pamphlet identifies the generating writer, since Koop's upright and independent demeanor becomes part of the persuasive process. All the information is conveyed in complete units and now the questions asked are not just "How can I avoid getting it from them?" There are no fragments. Even the bulleted statements under recommendations are complete sentences. While the different parts could probably be read at different times, the whole is clearly framed, by the personal foreword and the closing by the Surgeon General himself. This text recognizes the emotional forces involved in a reaction to AIDS, and tries to give an informed response to them, thus recognizing controversial issues, rather than trying either to deny or simplify them. The diction is at times polysyllabic ("promiscuity," "mutually faithful monogamous relationships") and at times scientific, at least in parenthetic definitions ("certain white blood cells [T-lymphocytes])." Here the scientific is in the parentheses, and the everyday diction is given the first place. Occasionally parenthetic definitions provided do not seem necessary (such as saying that "during" means "start to finish") and they usually then serve to define a term more specifically, rather than in slang or scientific terms. The scientific names of opportunistic infections ("Kaposi's sarcoma" and "pneumocystis carinii") are present. The illustrations include some far more technical than we have seen in earlier examples. There is a cross section of the body showing the location of the rectum, and a cut-away close-up of epithelial rectal tissue. Yet other drawings are non-technical -- a dirty intravenous needle, a condom package, a blood donor's arm, a cat, a mosquito. There is a diagram of the working of a virus, but only at the technical level of newspaper or news magazine graphics.
This spectrum of health education tones, styles, and implied audiences -- from thoughtless party-animals to worried citizens, lovers of precision -- gives us a context for appreciating the delicate choices made during the writing of the ANM. None of these models were exactly suited for a mailing to 108 million American homes. Only through the ANM drafting process was a new model realized, empathetic and professional, assuming intelligence but explaining carefully. The successes of the final draft, however, were only achievable at the end of a long process
Personalizing the Composing Process
To understand the composing process, a description of the magni-tude of the staff involved is necessary. Appropriations legislation for fiscal year 198819 included among its provisions a mandate that CDC shall "cause to be distributed... an AIDS mailer to every American Household.... by June 30, 1988,"20 and further specified that no review external to CDC was necessary.21 (A previous attempt at a national mailing had been tabled into non-existence by the Council of Domestic Policy Advisers). Appeals to a general audience are often thought to be the most vexed and difficult of health education genres. Particularly on a complicated subject, is it possible to answer the questions, so many different questions, from so many different people? In this case, a frequency rated list of calls to the AIDS National Hotline established what "the people" wanted to know. Focus groups in ten cities, community meetings in twenty-eight different cities and leadership forums for more than 600 AIDS leaders and health care workers had been done in October of 1987, in the interests of the larger campaign contract which Ogilvy & Mather already held, "America Responds to AIDS." Those specific questions also had to be addressed within a context of public fear and concern visible in the popular media. This general audience could reasonably be assessed as non-expert, fearful, and desiring distance from AIDS. The aim of the final draft was to change that audience into a knowledgeable, calm, and AIDS-related community. While specific behavior changes were, of course, desirable, the more proximate goal was the mainstreaming of a caring and fact-based attitude toward AIDS.
The ANM's composing team began this process by realizing how many different audiences could be involved and how differently those audiences might view the ANM. With the goal of reaching a large final audience, a number of smaller intermediate audiences were assembled. The process of composition lasted from December 1987 to March of 1988, when final copy went to the printers. At least sixteen numbered and differentiable drafts were produced during this period, in addition to additional revisions of format judged minor enough not to have been saved. Some members of the writing team speak of twenty to thirty drafts.22 The Atlanta office of Ogilvy & Mather and the staff of the National AIDS Information and Educations Program of CDC, in Atlanta, were the two largest groups involved on a daily basis, and CDC scientists and administrators, and the Surgeon General and his staff were frequent and significant contributors. The drafts produced by the exchanges between these two groups were responded to by twelve focus groups in January, embodying various age, race, and economic levels, from Danbury, Denver, Kansas City, Los Angeles, New York City and Raleigh. In February, a series of one-on-one interviews was conducted with a group described as "health educators." That term included twenty-one individuals working at the national, state or local levels with AIDS issues in their communities. There were also meetings in February with Surgeon General Koop and his staff in Washington. Health educators, researchers, CDC and PHS officials, blue collar and white collar, male and female, those who sought technical accuracy and those who wanted a comforting tone -- all became part of the process. The revising community both thought of themselves as an audience -- how did they, with particular expertises, react to a draft -- and thought of that other audience -- gauging its reactions even when quite different from their own. Thus as the document was revised, the importance of audience reaction was doubly present: revisions might be suggested that would please the proximate audience of one revising community but then eventually subtracted because of the needs of the remote final audience. Because each draft was divided into many separate sections, revision proceeded at different intensities in different sections. Some were intact almost from the first draft and only polished and pared; others were complicated and enlarged. Still others were pared down, then enlarged, then pared again. Academic models of revising (often with single authors and repeated start to finish perusals) did not apply; workplace models, more collaborative and less global in focus, did. Let us look now at some of those revisions.
This paper will review three sections in which the change from first to last drafts was sizable and interesting, and which, in somewhat different ways, show us the way in which a concern for audience reception is kept to the fore. They are: the first page covering letter for the brochure, the discussion of AIDS symptoms, and the quiz provided for readers.
The Covering Letter
The headline on the first draft first page reads "The Facts About AIDS" and the next line begins the address to the audience with a negative imperative: STOP. The appeal to the world of biomedical discourse, evident in the term "facts," is delivered in an authoritarian, Joe Friday tone. "This booklet is about AIDS. What it is. How it is spread. How to avoid it. How to help others avoid it." The final sentence of the section concludes with a reference to "facts": "We must make the facts very clear to all people." "We" is not defined personally, but only as the source of authority, the deliverer of "Important messages from the Public Health Service." Though the text speaks of the recipient's choice to open the sealed brochure, it is the authoritative "we" who decides to give the lowly "you" that choice.
By the final draft, the first page presents a much longer document, the message from Surgeon General Koop; for that reason it is useful also to look at the first draft's prototype of that message, though it was positioned in the first draft on the last page. The tone is determinedly optimistic. It promises answers, and discusses the complete elimination of risk, saying, "And you can do so without giving up meaningful and fulfilling intimacy, closeness, and companionship." Though the voice is given a personal attribution, its heartily simplifying tone reminds one of ads for miracle diets to "Lose weight while you sleep. Guaranteed."
By the final draft, the first page message from Surgeon General Koop avoids both the tones of military brusqueness or of overly assuring platitudes. The headline is now "Understanding AIDS." Interestingly, at the point in the process where focus groups first saw this title, one focus group reader almost plaintively sought the reassurance of a title with military implications. "It still says 'Understanding AIDS,' and as I said before, I'd like to see it say something else: 'Fighting AIDS' or 'How We Combat AIDS' or 'What Is Our Official Position'." "Fact" has disappeared entirely from the diction, succeeded by "information" and "issues." Though the implicit metaphor of warfare remains, in the reference to "AIDS, a health problem that the President has called 'Public Enemy Number One'," the soldiers, if you will, are a varied group: you, your family, and your loved ones. The imperatives now, in this draft, are for action, instead of for stopping action. "Discuss," "Know, read, get involved." The call to responsible behavior is not an imperative but an exhortation. ("I encourage you to practice responsible behavior based on understanding and strong personal values." "STOP" was directed to the audience in the first draft. Now, the word reappears in "to stop AIDS" or "Stopping AIDS is up to you." The audience has moved from the implied object, acted upon position, to the implied subject, or actor position. The message of the first draft emphasized distancing the audience from the disease. The final message asks for actual involvement. Instead of the simple polarity between the "we" of governmental authority and the "you" of the uninformed and unempowered audience, the final draft includes as well a "we" encompassing both categories: "We all must know about AIDS."
The Discussion of AIDS Symptoms
The changes in the sections discussing AIDS symptoms show a similar movement from the discourse of biomedical or epidemiological facts toward the discourse of the average reader. "They want to make sense of AIDS," noted one memo from the writing group, "not pass a medical exam." Early drafts seem to want to place as many "facts" as possible in the document; the later revisions show a desire to say enough to spur behavior change without giving an overload sufficient to induce confusion, despair, or false hopes. From a stage where the draft appeared to be an attempt to make it look knowledgeable to those who knew a great deal more, it evolved into a document designed to be comprehensible to those who knew a great deal less. The structure of the section addressing symptoms changed radically from the first to last drafts, and the tone and accompanying visual changed similarly. Initially there were eleven sentences and a list of symptoms introduced by bullets. Knowledgeable readers saw problems with it from the very beginning. The bulleted list is actually of ARC symptoms, yet it is the strongest visual element under the heading of "AIDS symptoms." When the passage gets around to describing AIDS-defining conditions, it is vague: "a rare kind of pneumonia," and "Other infections they cannot fight off," except in the case of naming Kaposi's sarcoma.
By the second draft, a very clearly divided discussion of the three stages of the infection is given. By draft four, the head now reads "What does someone with AIDS look like," but instead of focusing on how we can see the disease in others -- with all the potential for discrimination that implies -- it concentrates on how we might recognize the disease in ourselves. Both the health education professionals and non-professional focus groups are troubled and their reactions provide an example of how the best approach for risk communication is not simply providing more and more information. Sometimes a decision has to be made to control information. Explaining the three-stage progression, say the health educators, "may lead to more confusion and fear." The focus groups had trouble with the three-stage division, just as the health educators assumed that there would be. Members of the focus group reveal that this detailed information has left them more fearful. "Maybe they could have been a little more clear on that"... "If I get this thing.... will I die from it." While, on the one hand, the health educators want technical precision in certain parts of the language, ("Don't say 'pneumonia,' say 'Certain kinds of pneumonia.'" "Don't say 'drugs,' but say 'sharing needles'.") they seem frequently to foreground not their own knowledge, but what the audience will make of that knowledge, whether that means rushing to discrimination, or increasing fear among people whose behavior is not high-risk. They are aware of what people do not know ("Tell people they may have sores in their mouths they are unaware of,") and what they might do with an overload of information ("Change the references to prostitutes because it makes it appear as if only bad people get a bad disease." While the scientific voices are careful to moderate absolute claims, they understand as well how the lay audience is inclined to worry. The focus groups showed themselves most sceptical on the "truth" of what they were being told concerning transmission through insects, casual contact, bodily fluids and transfusions. The technical experts wanted to go as far as they could to calm fears, but were very aware of just where the boundaries of certainty were. Still, the technical experts' comments exhibit a patience with the limitations of their audience. Calls to the national AIDS hotline, for example, had revealed that a fear of contracting AIDS while donating blood was (and still remains) a major issue for callers. Technical commentators patiently repeat the need to calm this worry, and somehow manage to avoid a tone of exasperation about this basic scientific misperception. In the final draft, the section headed "What Does Someone With AIDS Look Like" contains no references to the three-stage nature of the disease, and the accompanying picture chosen is now Anthony S. Fauci, an NIH research scientist, who tells readers that they cannot tell from appearances, but that they are in no danger from non-risk-behavior contact. So, while the changes in the discussion of symptoms do not exactly parallel the direct movement toward a caring community that the page one changes do, the move is certainly away from a specificity of reference more associated with the world of "fact." The discussion has become more general, instead of including specific disease names. Yet, at the same time that the text has become less scientific, the choice of accompanying photo has become more so. The accompanying picture has shifted from a PWA to a volunteer, to a researcher, Anthony Fauci. Fauci's title at NIH is now actually longer than his comment, as if the subtext of "fact" is now not directly mentioned but invoked by his institutional identification, as if to comfort those who want certainty in a section which is insisting on complication and uncertainty. Does this change seem a decline, implying that the facts are just too complicated for the average person to understand? A look at the reactions to the symptoms section suggests that it is, in fact, not a disdain for the audience, but a strong awareness of the human cost of leaving in the three-stage model which makes the difference. If you bring the three-stage model out into the open, you are encouraging the classifying of human beings, like bugs or butterflies. Under that system, the audience is allowed to dominate someone else's problem with an intellectual grid. The final version pushes humility and uncertainty in the fact of a complex problem.
The Quiz
Finally, the quiz. The first draft version, in keeping with the general militaristic tone of draft 1, was headed with an imperative: "Test What You Know About AIDS." It included six questions, as did all subsequent versions. The areas were risk groups, testing, transmission, and relation to drug use. The only change in the content areas in subsequent drafts was the dropping of a question on mood-altering drugs. Initially, only one question had a true/false answer, some questions had multiple correct answers, and the answers were printed as upside down numbers, with no comment.
In the second draft, the headline became not an imperative but a question: "Do You Know Enough to Talk About AIDS? Try this test." Instead of a test for its own sake, now the usefulness of talking to others is emphasized. Of the five true/false questions, now all have false answers. The drug question has been replaced by "You can tell by looking..." By draft nine, "Test" in the head has been replaced by the much less-threatening "Quiz," and by the final version, all the true/false questions have true answers and all the answers have several sentence replies and cross references to other sections of the mailing. The major influence on the structure of the quiz here seems to have been from the professional copywriters, with the exception of changes in question three, the condom question, which both focus groups and health educators were united in disliking. Early versions were labeled by focus groups as "trick questions," and it is after the input of health educators, prior to draft six, that the wording was changed. The first version was "Condoms are most effective in preventing AIDS when they are made of -- choose one or several --latex, lambskin, or contain a spermicide such as nonoxynol-9." Draft two asks if they were "the most effective way to prevent the spread of the AIDS virus" and the answer was false, with "Not having sex with an infected person" given as the correct answer. "With an infected person" was deleted from the answer in later versions. Draft seven reads "Condoms are an effective way to prevent the spread of the AIDS virus," and now the answer was "True," followed by this commentary: "However, the most effective preventive measure against AIDS is not having sex or sharing drug needles." Emerging from the meetings in Surgeon General Koop's office is the more qualified statement, "Condoms are an effective but not foolproof way to prevent the spread of the AIDS virus." The only change in the answer is the final change to "shooting drugs" from "sharing drug needles." Clearly both the health educators, and the public health officials wanted to avoid undercutting the campaigns for condom use, at the same time as they did not want to invoke a risk-free universe. In the final draft, using condoms and "true" are associated in the public's mind, as part of the general emphasis on positive reinforcement, on what you do know, what you can do, instead of the earlier simpler world of "STOP."
Discussion
In these three areas alone -- and the drafts of the entire document provide a much richer and lengthier confirmation of this -- no one voice of the participating community of revision was allowed to predominate. The scientist's concern for precision of disease description was balanced by health educators' awareness of how an excess of information could mislead, disturb, or lead to discrimination. The judgmental, stentorian tone proposed by the first draft -- and it is only fair to mention that no one participating in making that first draft ever expected it to be the final draft -- was replaced by the fireside chat of the final draft. An abstract governmental proclamation was crafted into a personalized appeal.
The writers of the ANM could not give every audience in America what that audience wanted or needed. Prohibitions attached to its funding prevented any references which might have fit the definition of advocating homosexual behavior.23 Thus, for the gay community, the ANM could not have been the most specifically helpful document. The ANM managed to avoid, however, increasing the atmosphere of divisiveness and fear surrounding AIDS. It is the process itself -- the repeated recycling of the drafts through a deliberately diverse community -- which provides a successful model of planning for the audience adoption of a risk message.
Conclusions
The ANM will never be repeated, for a number of reasons. It cost upwards of $25,544,853.00, according to the Macro study estimate.24 Costs have also been expressed in terms of individual copies, between 20cents and 25cents cents per copy.25 The outcry against a similar sum of money being used again, to reach only the same general audience would be great, whether coming from those want money directed to research and treatment or from those who simply want less money spent in times of economic downturn.
The ANM will also not be duplicated because the need for it has gone. Lest this statement seem startling in the context of new AIDS deaths each day, consider how the fulfillment of "need" is measured for any risk communication document. The "need" for a document of risk communication is not defined as the disappearance of the risk. For the scientific community, the "need" for HIV research will stop only when no more new cases appear and we can treat effectively all those which do. For the general public, measuring the effectiveness of money spent on any kind of risk communicating, standards are often extreme. Either the risk disappears and therefore the money is well spent, or the risk was never perceived as such and therefore the money was poorly spent. For specialists in the field, however, our definition of need is narrower than either the scientific or the general community and involves the conjunction of form and audience. All America, said Congress, had to be told something about AIDS. And they were. Our national government had not directed a public address to every citizen, at a level designed to be read by everyone. That having been done once, has been done enough. What we know now, increasingly, about health education in the AIDS area, is that the more narrowly focused the audience community, the more effective the efforts will be. As we watch revision choices being made, we see that each step toward a larger audience may involve a step away from a more specifically focused group.26
It will not be done again, and it should not be done again. What, then, in a field where advances move so rapidly, are we doing looking at it at all? First, we can learn how the interaction between technical scientific experts and those with public communications skills works, that it can work, and that it can lead to a product rather than to a deadlock.27 We can learn the centrality of emphasis on education as a process. All the drafters were working with their different perceived audiences, rather than simply concentrating on the product. For risk communicators, the model is of non-competitive interaction between those with different constituencies. The scientists want technical accuracy, the ad writers want clarity and punch, and the health educators focus on the ways in which they know a text will be read, whether that is a "right" way or not. Does a major disease or environmental crisis make all concerned "make nice"? Hardly, as anyone who has ever staffed a press conference knows. Only an active conception of the audience's needs can blend the competing needs of scientists, writers and audience together. "Making sense of" the history of the ANM project offers a model of a positive interaction between human beings and complicated and clinical knowledge.
This study has attempted to look at the material of the ANM drafting and at the final product, and to look in two directions at the same time, toward that diverse group who produced it and that diverse group who were to read it. In both directions, this study has tried to show, we see not institutions or diseases, but, to borrow Evelyn Fox Keller's terminology:28
... embodied human actors without whom there would be neither language nor science. My starting proposition is self-evident: Science is a product of human actors engaged in material interactions with the objects they encounter, and attempting to craft those interactions into a way of making sense of the world -- especially the kind of sense that will foster the dual prospects of agency and control.
Neither writers envisioning an audience nor audiences facing a risk must lose sight of these central and energizing prospects -- agency and control. Given the restrictions upon its specificity and the size of the audience it was mandated to reach, that the ANM -- both in its process and in its product -- managed to maintain and enhance these two prospects is notable. To some extent, specialists and historians will, indeed, judge its success by the difficulty of the task and by what we can see of how it improved from its first attempts.
Its process of negotiated drafting is probably its most valuable lesson, and its attention to the audience, rather than to the competing truth claims of the various parties involved, was central to the success of that process. Not every risk communicator has the time to make sixteen drafts or involve as many individuals, but every risk communicator or planner can insist upon the centrality of an audience's needs in the midst of competing communities of certainty.
References
* Dr. Veeder is Associate Professor of English, Indiana University Northwest. She received her B.A. (English) from Saint Mary's College, and her M.A. and Ph.D. (English) from University of California, Berkeley.
1 Barbara Gerbert & Bryan Maguire, Public acceptance of the Surgeon General's brochure on AIDS, 104 PUB. HEALTH REPTS 130 (1989).
2 MACRO SYSTEMS, INC., FINAL REPORT: CASE STUDY OF NATIONAL AIDS MAILER UNDERSTANDING AIDS (1990).
3 Carol Ledbetter & Don Johnson, AIDS: Reading Level Analysis of Understanding AIDS, 4 AIDS & PUB. POL. J. 168 (1990).
4 Essential to cultural studies is its insistence that cultural studies in fact "has no distinct methodology, no unique statistical, ethnomethdological, or textual analysis to call its own. Its methodology, ambiguous from the beginning, could best be seen as a bricolage. Its choice of practice, that is, is pragmatic, strategic, and self-reflective" LAWRENCE GROSSBERG, CARY NELSON & PAULA TREICHLER, CULTURAL STUDIES 2 (1992). Cultural studies accounts draw "on multiple methods simultaneously -- meshing survey research with ethnography, for example, or information from modern marketing research with more utopian conceptions of empowered consumers." Id., at 3. Cultural studies embraces both "a broad, anthropological and a more narrowly humanistic conception of culture" Id., at 4.
Readers interested in, and unfamiliar with, a cultural studies approach may consult the following good beginning references: Douglas Crimp, Portraits of People With AIDS in CULTURAL STUDIES, supra, at 117; Cindy Patton, INVENTING AIDS (1990); Paula Treichler, AIDS, Gender, and Biomedical Discourse: Current Contests for Meaning, in AIDS: THE MEANING OF HISTORY 190 (Elizabeth Fee & Daniel M. Fox eds., 1988), SIMON WATNEY, POLICING DESIRE: PORNOGRAPHY, AIDS, AND THE MEDIA (1987.) See also, AIDS: CULTURAL ANALYSIS/CULTURAL ACTIVISM (Douglas Crimp ed. 1988).
5 Brenda Dervin, Mass Communicating: Changing Conceptions of the Audience in PUBLIC COMMUNICATIONS CAMPAIGNS, at 71 (Ronald Rice & William J. Paisley eds. 1981).
6 This term emerges in the work of Janz and Becker, 1984, and Strecher et al., 1986, as summarized by Mildred Seldes Solomon, Selecting Appropriate Strategies in EVALUATION AND EFFECTIVE RISK COMMUNICATION WORKSHOP PROCEEDINGS, 69 (Ann Fisher, Maria Pavlova & Vincent Covello eds. 1991).
7 OFFICE OF TECHNOLOGY ASSESSMENT, HOW EFFECTIVE IS AIDS EDUCATION?, 2 (1988).
AIDS education has two purposes. First, educational programs are intended to influence people to adopt or maintain behaviors that prevent HIV transmission. This purpose applies to AIDS education for all groups, including efforts directed to people with risky behaviors and to the general population, which consists mostly of people at low risk.... The second purpose of educational programs is to maintain and promote social cohesion, a goal that relates mainly to education for the general population... intended to relieve anxiety among people at little or no risk and to further the second goal.
Appendix B, at 95 characterizes the ANM as being designed to:
increase knowledge and influence attitudes and beliefs.... The goal of the campaign is to change people's attitudes and shift the environment (i.e., more compassion for AIDS patients), and through these attitude changes affect behavior.
8 Risk Communication Challenges, 11 RISK ANAL. 173 (1991).
9 This drafting material, now in the author's personal files, is available to interested scholars upon request. The material was obtained by direct letters of query to CDC's National AIDS Information and Education Program in Atlanta. Individuals at that office referred the writer to members of the writing team at Ogilvy & Mather's Atlanta office. Ogilvy & Mather provided copies of sixteen drafts of the English language version, and miscellaneous ancillary communications (internal memos, progress reports to CDC.) The writer was told that there was no additional material available. Certainly the Ogilvy & Mather files at the Washington office, for example, might have yielded a slightly different set of documents, but it was impossible to ascertain if every paper had been seen. Advertising agencies are dealing with many clients and a high volume of material. Their interest in maintaining archival material is not great. They maintained earlier versions of the product developed for the client, and individuals still working there were interested in discussing the process, but they were not accustomed to, or expecting to become, the subject of a study of the development of one campaign. Thus it might be argued that some of this material is random or anecdotal. Here again, the difference in perspective between humanistic and social scientific studies is evident. In studying the drafting and revisions, for example, of a famous writer, humanities scholars do not assume that the "completeness" of what they are considering must be scientifically demonstrated. The humanities scholar is trained to infer from the material available, and see if these inferences seem to present consistent or notable patterns over the course of the project. What might seem risky to a social scientist seems standard operating procedure to a student of the humanities.
All conclusions drawn here are based on the English language version, and not the Spanish version.
Special thanks to Fred Kroger, Lynn Herring, Mark Rosenberg of CDC, and to Stephen Heller of Ogilvy & Mather.
10 For additional description, see the section following the discussion of three contemporary models.
11 As earlier notes have indicated, principle areas of difference may involve methods of data collection, a reliance on inferential analysis, and a decision to analyze the text as a series of choices made by writers who are aware of many variables: scientific data, audience self-definitions, audience fear, and the standards of effective public information advertising. This study looks at a variety of message senders, approaching a variety of message users, with approaches which become, across the progress of the drafts toward final copy, more various and flexible. This study combines rhetorical analysis, what might be called attempts at retrospective ethnography, and a sensitivity to the emotionally and politically charged valences of the AIDS material.
12 Supra note 6, at 197.
13 INDIANA STATE BOARD OF HEALTH, WHEN THE PARTY'S OVER... (1988).
14 AIDS: NOW IT'S EVERYONE'S CONCERN (1988). The booklet indicates that it was developed in cooperation with the Citizens AIDS Project and the American Pharmaceutical Association and supported by an educational grant from Burroughs Wellcome Co.
15 U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES, SURGEON GENERAL'S REPORT ON ACQUIRED IMMUNE DEFICIENCY SYNDROME (1986).
In a way more characteristic of a cultural studies approach than of a social scientific data gathering event, the pieces were assembled on the basis of what would be visible to the general public in the course of their daily lives, within the region in which the author, a member of the general public, received a mailed copy of the ANM. Schools, public libraries, pharmacies, and discount stores often made material available. Because these sites were by definition reaching to non-specialist audiences, the writer chose to consider their choices as a significant sample. Of the fifteen pieces easily available within a brief period of time, the three pieces considered here were chosen as representative of the major approaches encountered. General assumptions about audience were able to be check in the case of the pharmacy pamphlet and the Surgeon General's pamphlet, but queries were not productive on the Indiana AIDS pamphlet. It is the contention of this study, however, that the assumed audience perceived by the general public is also a significant factor.
16 Supra note 13.
17 Supra note 14.
18 Supra note 15.
19 See, e.g., Pub. L. No. 100-202, 101 Stat. 1329 (1987).
20 101 Stat. 1329-265.
21 Id. See also, 133 CONG. REC. (daily ed. Dec. 22, 1987) (Part II).
22 Some of those additional drafts may have been of minor boilerplate revisions, not saved separately. Or, the writers may have, in fact, made more versions in their memory than in fact. Or, there may be some combination of both these explanations.
23 133 CONG. REC. H12892 (daily ed. Dec. 22, 1987). Section 514(a).
Notwithstanding the matter under the heading "CENTERS FOR DISEASE CONTROL", none of the funds made available under this Act to the Centers for Disease Control shall be used to provide AIDS education, information, or prevention materials and activities that promote or encourage, directly, homosexual sexual activities.
24 Supra note 2, at 62.
25 Supra note 2, at 64.
26 JEFFREY A. KELLY & JANET S. ST. LAWRENCE, THE AIDS HEALTH CRISIS: PSYCHOLOGICAL AND SOCIAL INTERVENTIONS (1988), especially Chapter 4: Behavioral Interventions. The focus of successful interventions, as Kelley and St. Lawrence state, is material which is specific in its focus on changes in practice, which is narrow on its focus within risk groups and communities, which is distributed through community-based programs, and which is not restricted solely to print-based materials.
27 Additional useful discussions of AIDS and public policy are in 4(1) AIDS & PUB. POL. J. (1989), and in AIDS: SEXUAL BEHAVIOR AND INTRAVENOUS DRUG USE. (Charles F. Turner, Heather G. Miller & Lincoln E. Moses eds. 1989).
28 Between Language and Science: The Question of Directed Mutation in Molecular Genetics in SECRETS OF LIFE, SECRETS OF DEATH: ESSAYS ON LANGUAGE, GENDER AND SCIENCE 180 1992).