"Les mots exacts or ‘saying it like it really is’ " (ISASI forum 28:1, p 22, March 1995) raised the issue of misuse or imprecise use of language related to accident reports. I was gratified to learn that others are addressing this issue. "Les mots . . " however addresses only one aspect of the general area of problem words. My experience with quality control problems attributable to words and phrases used in investigation reports or other investigation work products affirms the misuse and imprecision issue and other problems, and indicates "problem words" and language should get more attention than they have received thus far from ISASI’s membership.

My earliest awareness of problem words for investigators grew out of observed problems during my research into the application of logic to control investigation process quality . Numerous investigations I have managed, participated in, observed or critically analyzed have provided me with insights into this issue and a list of problem words that other investigators might find useful.

Words say something.

Investigators produce their work products for others to use. Users may range from investigators’ supervisors and managers or reviewers to anyone in or associated with the aviation industry, such as pilots, flight crews, mechanics, instructors, traffic controllers, equipment designers, procedures manual writers, insurance claims managers, regulatory personnel, law enforcement personnel, emergency responders, research professionals, and safety or risk management personnel. Uses range from aviation system design to aviation system training and operations to regulatory initiatives and settlement of claims, among many others. Users of investigation work products are investigators’ "customers." I have approached this discussion from the perspective of the investigator’s "customers."

Investigators have in mind what they want to communicate to their customers when they write about what happened during an accident or incident. Sometimes their written words communicate unintended messages to their customers. The purposes of this paper are to heighten awareness of one category of problem words and phrases in investigation work products, describe the problems they introduce, and share ways to prevent these problems.

More examples of problem words exist than could be covered adequately in one article. I will treat one type here, and others in future papers. Each paper will examine one or more words or phrases of the kind often found in investigator’s work products, and discuss problems with their use, and indicate ways to solve the problems.

Categories of problem words.

Certain words reflect or introduce problems with an investigation. Problem words fall into several categories, including

1. misused words that reflect an investigator’s logic difficulties, addressed by the "Les mots . . " paper.
2. comparison problem words that reflect an investigator’s unstated or unsupported assumptions, biases, or subjective judgments, and therefore convey unjustifiable conclusions.
3. confusing words that prevent users from visualizing what happened.
4. compound words that force a user to infer or jump to conclusions about who did what.
5. contextual dependent words that say exactly what the investigator intended but the investigator does not describe their context, permitting users to read them in a different context.
6. cover-up words that an investigator chooses deliberately to mask incomplete or deficient investigations.

Comparison Problem Words: Comparative adjectives.

A "comparative adjective" is a word that investigators use to convey the notion that something did not happen as they think it should have happened. This problem category can be illustrated, for purposes of this discussion, by referring to the article "Improper Maintenance induces Hazards in Piper PA-31-350" on page 24 in the same issue of the forum as the "Les mots . " paper. The forum article shows both the problem word and how the problems it reflects can be overcome.

The comparative adjective problem is illustrated in the forum article’s second paragraph. "Safety (lap) belts had been installed improperly on several of the passenger seats - - " Improperly is a comparative adjective, reflecting a conclusion by the investigator. If no supporting basis for the conclusion is reported, it is a subjective and unsupported conclusion which can not be verified or validated within the report for quality control purposes. Without supporting data, the conclusion requires a leap of faith in the investigator to be accepted. Reports with subjective conclusions are inevitably more vulnerable to controversy - individuals accused directly or implicitly by investigators of erring are not likely to display much faith in the investigator’s subjective judgment.

In many small-scale investigations or investigation of selected aspects of an occurrence, particularly those conducted by one experienced individual, the investigator concludes action or performance was "improper" for one of two reasons:

Reason 1: the investigator "knows" from personal experience what the person should have done, and assumes (concludes) that because the incident or accident occurred, the person didn’t do it "properly."

Reason 2: the investigator doesn’t know how something should have been done, and just assumes (concludes) it was done "improperly" because the incident/accident occurred (worst case.)

Solving the comparative adjective problem.

The solution to the comparative adjective problem is to identify and report the specific standard(s) for determining the "proper" (or expected) action, then describe how the standard(s) were communicated, and then report what the person or object did during the occurrence. Let the comparison and contrast speak for itself.

The article indicates the standard for comparison was a maintenance manual, but it does not state the standard completely or unambiguously. It does not state that this maintenance manual information was communicated, nor does it show that the manual would have been all that was necessary and sufficient to achieve the desired (proper) actions.

Sometimes investigators find that comparative adjectives like "improperly" must be used to satisfy some organizational policy or Manager’s demand. When that occurs, the investigator should take the initiative to present the standard for comparison and its communication, completely and unambiguously. This need is especially significant in incidents involving discussions of "improper" human actions.

On the other hand, investigator’s supervisors should ask for the behavior standard(s) which a specific individual was "programmed" to implement, and how both the investigator and the individual involved were supposed to know what that was and when it was applicable. Supervisors may hear that the standards for comparison are technically difficult or time consuming to identify or acquire. The same may be said for objects which did not perform "properly" and are blamed for an occurrence, although the expectations may involve investigating different kinds of sources.

Flagging problem words.

Comparative adjectives, such as improperly, poorly, inadequately, incorrectly, substandard or mistakenly, should be red flags for investigation work product users (and investigators as well.) These words should stir users to demand an unambiguous description of all the relevant standard(s) used to compare what happened with what the investigator thought should have happened. The process is somewhat comparable to preparing a citation for violating a law or regulation: first one must define what regulation was violated, then describe who or what did what and when to have violated that regulation.

I have found that continuing challenges to unsupported comparative adjectives by report reviewers will quickly motivate investigators to discontinue their use, which will improve the quality and value of their investigations and reports.

(Next issue: comparative accusations ("failed to . . . " and its twin "did not . . . ")

Submitted for publication in the ISASI forum on 6/7/95