One important aspect of technical writing is the production and use of procedures. Though technical writing serves a variety of purposes, teaching, informing, persuading, and even questioning, one of its primary and most common purposes is the "how-to" function of providing procedures. There is a great deal of information available on writing procedures, the vast majority of it focusing on software documentation and product documentation. This paper, however, focuses on procedures for hazardous environments or situations. Specifically, this paper will discuss the rhetoric used in communication of hazards inherent in procedures or in procedural consequences, and how rhetorical theory and practice can help an author to assist the reader in avoiding mishaps.

Procedures are used extensively in some fields, and to a lesser extent in nearly every field of endeavor. Whether delivered orally, in written form, or developed from experience, most everything we do involves procedures. We may not always look at a cookbook when preparing a meal, but we had a recipe (written procedure) or an instructor (oral procedure) the first time we prepared a meal. More experienced cooks don't need written or oral procedures, but the new menu items they come up with are based, in part, on standard procedures they use repeatedly without even thinking about it.

Most people, when they think about procedures, think about the booklet that comes with a set of shelves from the local super mart, that one that most men seem to refer to only as a last resort when if can't figure out how it fits together. People tend to think about the cookbook described above, or about the books never looked at in the packaging of the last software program purchased. Even in the professional and pedagogical society of technical writers, most discussion of procedures seems to be centered on software documentation or product documentation. This is fine for most technical writers in these communities, but it completely ignores a whole genre of procedures, those involving safety or life-threatening issues.

In the military, everything is done by procedure. There is a procedure for using a dishwasher, for cleaning a rifle, for driving a tank, and for every one of the thousands of daily actions necessary to operate an aircraft carrier at sea. Every one of these procedures is there for a reason. They were developed over time, and designed to avoid the deadly or expensive mistakes that would be likely or certain to occur if these procedures did not exist.

During the Vietnam War there was a fire on the aircraft carrier USS Forrestal (US Navy, video). A live missile from one plane was accidentally launched, flying across the deck and striking another plane that was lined up waiting to take off. Instantly the plane struck by the missile burst into flames and spilled burning fuel on the deck, igniting the planes on either side of it. Rescuers rushing to help the pilots escape the cockpits of the burning planes were killed, along with some pilots, as the bombs on the planes were set off by the heat. The explosions blew holes in the deck allowing burning fuel to penetrate to the decks below, causing more deaths and threatening to spread fire throughout the ship. Men below decks screamed as the burning fuel engulfed them or trapped them in tight quarters.

All hands were called on to fight this fire. Men on deck began to fight the fire with hoses that sprayed foam to smother the burning fuel. Others, trying to help, brought fire hoses from below decks and turned them on the fire. These hoses, designed for general use and not oil fires, sprayed water instead of foam. Since fuel is lighter than water, the burning fuel just floated on the water, which helped to carry the flames below deck. Worse, the water began washing away the foam that had already been applied, allowing those areas to reignite.

Below decks the fire was spreading, the heat and thick smoke made it impossible to fight this fire without breathing apparatus and protective clothing. Some of those trained in the use of the breathing devices were among the rescuers killed in the first minutes of the fire. Others were trapped in the compartments that were burning. Men struggled to figure out how to don the gear so they could help with the firefighting efforts, losing precious time as the flame continued to envelope more and more of the ship.

Eventually, this fire was controlled. Someone on deck with enough authority to make others listen figured out that the water hoses were doing more harm than good, backing those teams away allowed the foam to do its job. Below decks, as the first firefighters had to retreat due to using up their air or heat exhaustion, they were tasked with helping the next shift to don protective gear. Slowly, one compartment at a time, they battled the fire until it was extinguished, 13 hours after it started. Tragically, it was too late for many victims of the fire, both on deck and below. After the fire was out and they were able to take stock of their losses, 134 men were dead with many more injured, "twenty-one aircraft were destroyed and 43 others were damaged. The cost to repair the carrier was reckoned at $72 million." (USS Forrestal Website)

Every new sailor is now required to watch a film of this fire. As they watch, instructors point out the mistakes, where procedures were not followed or did not exist. Had proper procedures been followed, many deaths could have been avoided. Procedures could have avoided the launch of the first missile, could have had the plane pointing away from the others, sending the missile harmlessly overboard, could have allowed the first teams to fight the fire more effectively, thus minimizing the spread and damage below decks. Not all of the procedures that could have helped with this casualty existed before it happened. Some were written based on the lessons learned from this incident, thus preventing future repetition of the same mistakes. That is one purpose of written procedures, they are designed to help avoid repeating others' mistakes. A person using a procedure not only learns what to do in order to accomplish a task, he also learns what not to do.

Three Mile Island and Chernobyl became common household names because of disasters that could have been prevented or minimized by following procedures. "…worker injuries during ordinary maintenance account for more accidents in nuclear power plants than system failures or communication problems in the control room." (Sauer, 307).

I was called on, recently, to work with an engineer to revise the maintenance procedures for the cranes at my work. The old procedures had worked adequately for twenty years, but failed to require inspection of one critical component. This component, a simple block of metal (called a "dog") that dropped down when the crane rail was not aligned with another rail, broke off after twenty years of use, allowing a crane to run off the end of its rail and fall to the floor 40 feet below, killing the operator and just missing a number of people on the ground. Several procedures, written since that occurrence based on the lessons learned, are working to prevent a repetition of this or similar incidents. New maintenance procedures require technicians to check the dogs for damage on a regular basis. New operating procedures prohibit the operators from routinely hitting the dogs with the crane at full speed, and require an inspection if they do it inadvertently. These new procedures enhance safety at all plants owned by this company, not just the one affected by the incident.

In the Navy I witnessed an explosion in the engine room of a nuclear submarine. A series of drills was being conducted, in which an electric motor was placed in service, run through a series of heavy-duty operations, then removed from service by disconnecting the clutch that tied it in with the still-operating (steam driven) machinery. When disengaging the clutch, the procedure called for the operator to hold the "Disengage" switch for 30 seconds after the "Disengaged" light came on, to ensure the clutch was fully retracted and locked in place. A new equipment modification was planned to install a 30-second time delay so the light would not come on until it was safe to release the switch. A new procedure, without the requirement to wait 30 seconds, was sent out to all of the submarines in this class with instructions to insert it into the manual when the modification was complete. The ship's librarian, having no idea if the modification was done or not but assuming it was since he had the procedure in hand, inserted the new procedure into the operating manual, thus effectively deleting the instruction to wait 30 seconds. That was the only difference in the procedure, and there was no visible difference between the modified and unmodified equipment. Because the operator was following the wrong procedure, the clutch was not fully disengaged, allowing the two halves to rub against each other. Analysts later said the discoloration of the metal adjacent to the clutch indicated that it had reached temperatures around 1400 °F. The clutch oil, which was normally not very flammable but did have a flash point around 1100 °F, ignited and caused the explosion. All of this was the result of a single step left out of the procedure. Fortunately, though the damage to equipment was extensive and the mission of the submarine had to be cancelled, no one was injured in this mishap.

Those called on to write procedures for hazardous operations have a specific moral and legal duty to help the reader prevent mishaps, that is a main purpose of procedures written for hazardous situations. "The law requires manufacturers to furnish complete directions for the safe and effective use of products that can cause harm." (Helyar, 128). In a Society for Technical Communication annual conference presentation titled "Designing for the Web: Special Consideration for Safety Information," Lisa A. Tallman said:

Designing safety information […] is an iterative process that involves knowledge of hazard analysis methods, legal standards and regulations, and consumer behavior. […] As technical communicators, we have the responsibility to make sure that the instructions and warnings that we write meet government and industry standards and provide [users or readers] with the necessary safety information at the right time. (Tallman, 2)

For portability, critical procedures are generally produced and used in a paper format, so this discussion will focus on printed procedural documents.

Procedural Discourse

In his article, "The Logical and Rhetorical Construction of Procedural Discourse," David K. Farkas gives this description:

Procedural discourse refers to written and spoken discourse that guides people in performing a task—in other words, its "how to" communication. […] If human beings did not set about to accomplish tasks (in other words, to change things), we would not generate procedural discourse. Thus, procedural discourse is largely about telling someone who is in one set of circumstances how to transition to another set. (Farkas, 42)

He goes on to describe four system states, or sets of circumstances, and three types of actions that cause systems to transition from one state to another. The four states are desired state, prerequisite state, interim states, and unwanted states. The three types of actions are human actions, system actions, and external events (42-43).

Farkas says that procedures are inherently rhetorical and exist in a social context, in that they are a means of "communicating with others and guiding them (or hoping to guide them) through a task." (Farkas, 43). He goes on to say that procedures must establish their own credibility, so that the user believes it comes from a "knowledgeable and trustworthy source." (43). For the documents we are discussing, which are generally issued through an authoritative source, such as a product manufacturer or a corporate maintenance system, this is not usually a problem. Poorly written documents, however, or ones that don't seem to apply very well to the situation of the reader, can easily lose their credibility despite this authoritative source.

A good deal of Farkas' article discusses "Streamlined Step" procedures, which have the following defining characteristics (45):

• Steps are brief.
• Format is simple.
• Action statements begin with an imperative verb.
• A title and little else preceding the steps.
• When online, hypertext is used to "layer" the information (though Farkas doesn't mention it, this function can be performed by a glossary, footnotes, or cross-references in printed text).

The brevity and simple formatting of streamlined step procedures makes them efficient and the most common form of procedure, both online and in printed procedures. They are relatively easy to produce and customize. This format, and its variants in the step-action table and step-action-response table are what I use most in my work. They are especially beneficial when the procedure is simple or when the expected user is experienced and doesn't need a lot of detail in order to accomplish the task. They don't work very well, however, for complex tasks being done by unfamiliar workers or for tutorials. The required brevity works against these uses.

There are many other types of procedures and techniques that are effective in writing procedures. This paper makes no attempt to explore all of them. Instead, we will now focus on the reader or user of the procedures.

Audience

David Goodwin, in his article "Emplotting the Reader: Motivation and Technical Documentation," points out that "a manual must emplot the reader, that is, must create an action-oriented role within a storyline that transforms the reader from a hesitant, if not reluctant neophyte, into a competent software user…" (Goodwin, 99-100). He says that audience motivation is an important part of effective technical communication. Although he is talking here about software manuals, these same concepts can be applied to most types of procedures.

Goodwin goes on to describe three types of readers involved in a technical writer's audience. The actual or real reader is the human being who buys, opens, and reads any printed materials. "First and foremost, then, the actual reader is a concrete reality. He or she has specific habits, preferences, knowledge, and beliefs." (100) The authorial or implied reader, on the other hand, is a construction of the author. This reader exists only within the text.

As writers, we shape the actual audience's responses by inviting them to recall what the authorial reader is expected to know, to accept what the authorial reader is expected to believe, to become what the authorial reader is expected to become, and thus to read our writing in a manner appropriate and consistent with its genre and purposes.[…] the union of the authorial with the real audience—and, in particular, a union that promotes a good reception of the text, should be paramount in the mind of the technical writer. (101).

The ideal reader, according to Goodwin, is a type of authorial reader who "…would always understand the intentions, as well as carry out the instructions, of the technical writer without fault or misapprehension." (101)

There is one more type of reader that I would add to Goodwin's list, especially in procedures involving safety issues. On my desk there is a sign that says "Every time engineers develop a system that is truly idiot-proof, someone comes up with a better idiot." I often mentally change the words "…engineers develop a system…" to "…I write a procedure…." I'm not saying that we should write safety procedures tailored to the most idiotic reader who could possibly come across it, but when safety is an issue we need to keep that reader in mind. That reader, then, would be the worst-case reader, the one most likely to misunderstand the intent or meaning of a step in the procedure, to skim over it without careful study, to do things out of sequence or to take shortcuts with the procedure.

Goodwin continues:

Actual readers, however, frequently resist authorial roles, very often because they resent, disagree with, or feel confused by the ideals inherent in those roles. Thus, if the authorial-readership role is demeaning, inappropriate, ill-defined, or in some other way alienating to the actual readership, a schism develops between the aims of the writer and the actions of the reader.[…] Pushed hard enough, audience resistance gives way to hostility […]. Hostile readers become likely candidates for one of two other audience roles: the excluded or the enforced reader (101-102).

Excluded readers are those who don't feel like a welcome part of the writer's audience. This could be because the author used all gender-specific pronouns that excluded the reader, or something else within the text makes the reader feel like a Republican at a Democratic convention.

Those who write safety-related procedures are well aware that, quite often, their readers are enforced readers, they have to be there, to make their way through the text regardless of interest or inclination. Where safety is an issue, it's extremely important to motivate the reader. Goodwin says that writers must provide incentives to the reader, to make them want to be the authorial or ideal reader and to accomplish the tasks presented within the text. He suggests several tactics for this, including casting the author as a hero, the use of visualizations, and various time-sequencing tactics. Most of these seem well-suited for tutorials, but difficult to apply to a sequential procedure without adding significant length to it.

In "Understanding Readers," Janice C. Redish discusses the reader in a business environment:

The most salient characteristic of readers in the workplace is that they are busy and don't have time to waste on dense documents. […] Even sophisticated, technical readers of technical documents want to be able to open a document, find what they need quickly, understand it easily, and have the most important information stand out visually on the page or computer screen. Readers are continually deciding […] whether more time and effort with a document is worth the additional benefit of learning or understanding. I've found it useful to think of this as satisfacting, to borrow a term from the work of well-known economist Herbert Simon (1976). Simon used satisficing to describe what administrators do when they "look for a course of action that is satisfactory or ‘good enough.'" They act without exploring all the options. They use "rules of thumb that do not make impossible demands upon their capacity for thought" (Simon, 1976). […] In many cases, satisficing is the only way that readers can cope with their work environment. […] in the workplace, people most often "read to do" reather than "read to learn" […] Busy readers don't generally study documents. Instead, they scan them to find the critical information and to act on that information. (Redish, 3-4).

Redish lists three things we can do to help busy readers get the information they need quickly from documents or procedures (see Redish, pages 6-7):

1. Provide good access tools. Careful use of table of contents, topical headings, indexes, and context make it easy to find the information needed.
2. "Design pages that can be skimmed, scanned, and searched easily." I have seen procedures where the action verb for each step was either placed at the beginning of the step or printed in bold letters, or both, thus allowing someone to scan quickly to the most relevant information if they were already familiar with the procedure.
3. For "read to do" procedures, use a structure that promotes action, such as numbered lists of steps.

In procedures involving hazards, it is more critical than any other type of procedure to know and understand the reader. This calls for careful cooperation between the writer, system engineers, and experienced operators or technicians familiar with the hazards likely to be encountered during operation or maintenance, the skills and culture of the people likely to do the tasks, and the context that will be present at the time. Redish discusses various models of communication in her article, and points out that both the author and reader interpret text based on their own experiences (see 7-11). Fortunately, individuals share many elements of their schemata, or ways of looking at things, with others in their specific culture. "Successful documents are those that make explicit connections to reader' prior knowledge and expectations." (12)

Understanding and perception of risk

In her article, "Health and Safety Information for Specialized Vocational Audiences," Kris S. Freeman says "The creators of effective health and safety materials and campaigns must also understand and address attitudes that impair workers' perception of their risk for disease and injury, and workplace norms that limit the acceptable remedies to reduce risk." (Freeman, 1).

According to Freeman, there are specific trends that have been clinically shown in regards to individual perceptions of risk and motivation to do something about that risk. Workers, especially those who work regularly under hazardous conditions, tend to feel "bullet proof" or immune to the danger (3). But it's not enough to convince these individuals of the danger. "People are motivated to change their lives only when they are personally at risk and [convinced] that the risk is serious. In addition, they have to believe that there are effective ways to reduce the risk, that the benefits of these remedies exceed their cost, and that they are capable of making the recommended changes." (3)

It is also essential that writers understand the workplace attitudes and constraints that might limit implementation of safety measures. Tight deadlines, incentives for taking shortcuts, and other cultural influences can all cause pressures that will lead a worker to resist or neglect safety precautions. (4). I cannot count the times I have seen people take unnecessary risks because of pressure to complete a task quickly or because the safety requirements were more work than the procedure itself. By motivating the reader, you maximize the safety of that reader, but effective motivation demands understanding.

Freeman gives these suggestions for maximizing credibility and effectiveness in promulgating safety information in the workplace (5):

• Visit the workplace several times and examine conditions personally. (5).
• Chose credible sources. Information resources such as trade journals or industry associations, local radio stations and a look at what publications are in the break areas will help to indicate what sources the workers find credible. (5).
• Study the culture and language. Discuss the work with individuals at various levels of authority. Participate in the work, and learn the correct terminology. "Using an inaccurate term for a piece of equipment or procedure can dramatically reduce the credibility of your materials." (5)

In nearly every safety situation, there are codes and regulations that apply. These codes cover everything from the color and appearance of safety symbols used in a document (ANSI Z535.3-2002) to required safety inspections for various pieces of equipment, environmental contamination limits, types of ladders that can be used to accomplish specific tasks, and the spacing of bolts in mine ceilings. For liability and safety reasons, it is critical that the procedure writer knows these rules and ensures that each procedure covers all legal requirements.

The article "Products Liability: Meeting Legal Standards for Adequate Instructions" summarizes with these suggestions for technical communicators (Helyar, 142):

• "Describe the products' functions and limitations;"
• "Fully instruct on all aspects of product ownership;"
• "Contain clear, correct, and tested instructions;"
• "Use words and graphics that suit the intended audience;"
• "Appropriately warn of product hazards;"
• "Offset claims of product safety in advertising or other materials with intense warnings;"
• "Present important directions or warnings so consumers spot and follow them;"
• "Meet government, industry, and company standards;"
• "Reach product users; and"
• "Timely inform consumers of defects discovered after marketing."

Feedback

One thing I haven't seen mentioned, but that is critical when writing safety-related procedures, is to give your readers a means of checking their own progress, of knowing they are on the right track. In a way, you are giving them feedback on their progress. Computing procedures often do this with screen capture displays, so the user can verify that their operation produces a screen that matches the illustration. Whenever there is a simple way to verify proper operation, it should be mentioned. Some examples are:

• Note the sound of steam as you open the valve.
• The conveyor belt will stop and the power light go out when you open the breaker.
• Check for voltage at the exposed terminal, then verify that your voltage instrument is working by checking a known live circuit of the same capacity.
• Watch the pressure gage and verify that pressure begins to drop.
• The following message will appear on the screen…
• Turn right at the third light. You will be heading north on Center Street.
• Use a micrometer to verify that clearance is no greater than 0.003 inches.
• If you hear a rush of water in the system, stop immediately and reposition the last valve operated to its prior position. Return to step X and repeat the procedure from that point.

In his chapter on Motivation, David M. DeJoy lists 4 elements that should be present in effective warnings. These are (DeJoy, 228):

1. "A signal word to convey the seriousness of the hazard or threat." Beginning a statment with "Warning!" or "Caution!" immediately lets everyone know that this is a hazard statement.
2. "A hazard statement that describes the nature of the hazard or threat." This could be a phrase such as "Electrical Shock Hazard!" or "Slippery floor when wet!"
3. "A consequences statement that describes what might happen if the warning is ignored." e.g. "Damage to equipment will result if…", "Severe injury or death is possible!"
4. "Instructions about the action or actions that will reduce or eliminate the threat." It's not enough to warn the user, you have to tell them what to do to prevent harm: "Use respiratory protection at all times when the paint booth is in use and for 15 minutes after the last painting operation"; "Wear chemical protective clothing to include, at a minimum, goggles and a protective face shield, apron, chemical-resistant gloves, and plastic arm sleeves"; "Do not enter!"; "Use proper lockout procedures to ensure this equipment is not energized during maintenance."

Conclusion

Technical communicators have a critical task in preparing procedures involving safety issues. When a product, situation, or operation is fraught with hazards, the warnings must be selective so that the more critical items do not get buried by trivial or unlikely items(Helyar, 135) Too many warnings can cause a reader to ignore all of them.

In order to be effective, each warning must be presented in a manner that conveys the actual risk to the reader, and accompanied with instructions on how to avoid or minimize the hazard. (Freeman, 2 and DeJoy, 228).

Technical writers need to be familiar with the situation, culture, and experience of likely readers. Using the wrong or an unusual term for an object can identify the writer as someone outside the culture and destroy credibility. Small failures in communication in seemingly ordinary circumstances can result in "gross consequences" far out of proportion to the effort that would have been required to avoid problems (Sauer, 307).

Hazardous situations and operations are an inevitable part of life and business, but there are always things that can be done to minimize the risks. With the state of litigation today, it only makes sense for a business to do everything possible to ensure a safe work environment. Because of their specialized communication skills and experience with motivation, technical writers are a key part of any effective safety program.

Works Cited

American National Standards Institute. July 25, 2002.

DeJoy, David M. "Motivation." In Warnings and Risk Communication, edited by Michael S. Wogalter, David M. DeJoy, and Kenneth R. Laughery. Taylor & Francis, Inc., Philadelphia, PA, 1999, 221-243

Farkas, David K. "The Logical and Rhetorical Construction of Procedural Discourse." Technical Communication, 46, no. 1: 42-53.

Freeman, Kris S. "Health and Safety Information for Specialized Vocational Audiences." In 48th Annual Conference of the Society for Technical Communication. Society for Technical Communication, Chicago. 2001. 537-542.

Goodwin, David. "Emplotting the Reader: Motivation and Technical Documentation." Journal of Technical Writing and Communication, 1991. 21, no. 2:99-115.

Gopen, G.D. and Swan, J.A. "The Science of Scientific Writing." American Scientist 1990 78:550-558.

Helyar, Pamela S. "Products Liability: Meeting Legal Standards for Adequate Instructions." Journal of Technical Writing and Communication, 1992. 22, no. 2:125-147.

Horowitz, Don (editor). Situation Critical: The USS Forrestal, video, 1997

Redish, Janice C. "Understanding Readers." In Techniques for Technical Communicators. Carol M. Barnum and Saul Carliner. New York: Macmillian, 1993. 15-41 (pages are numbered 1-23 in reprint copy that I have, these are the page numbers referenced in this document).

Sauer, Beverly A. "Communicating Risk in a Cross-Cultural Context." Journal of Business and Technical Communication, Sage Publications, Inc. 3 July 1996, 10, no. 3:306-329.

Schneider, Barbara. "Clarity in Context: Rethinking Misunderstanding." Technical Communication, 49, no. 2:210-218.

Simon, H. Administrative Behavior (3^rd ed.). New York: Macmillian (1976)

Tallman, Lisa A. "Designing for the Web: Special Considerations for Safety Information," In 48th Annual Conference of the Society for Technical Communication. Society for Technical Communication, Chicago. 2001.

US Navy, Trial by Fire. Documentary on the USS Forrestal fire. 1973

USS Forrestal Memorial Website, http://www.viethero.com/Tribute/Forrestal.html

USS Forrestal Website, Fire Page, http://www.forrestal.org/fidfacts/page14.htm

Wright, P. "Writing Technical Information." In E.Z. Rothkopf (E.), Review of research in education (Vol 14:327-385). Washington, DC:American Educational Research Association. 1987 (as quoted in Redish, 5).