Avoid metaphors with exclusionary overtones

What it is

An exclusionary-metaphor review asks whether a metaphor helps the intended audience understand the current task or instead relies on narrow cultural experience, demeans people, minimizes real harm, or misdescribes system behavior. The response is not to ban every metaphor. Key tasks receive a direct statement of object, action, and outcome; a retained metaphor is a validated aid rather than the only key for decoding the task.

Why it happens

A metaphor borrows a familiar source domain to compress a complex idea, so shared experience can reduce learning effort. Without that experience, users must first decode lore, sport, war, or bodily imagery before understanding the product. Metaphors also carry roles and value relations: calling an error “crippled” or an easy mode “dummy mode” puts real people into a negative slot, while game-like language for data loss can obscure consequences. Direct wording establishes a minimum shared meaning; a metaphor must then be evaluated through comprehension, emotional impact, and task risk together.

Studying it

Test candidate metaphors inside complete messages and tasks rather than asking whether participants like an isolated word. Invite people with varied relevant experience to explain the product object, expected action, and consequence, recording misinterpretation, inability to explain, offense, or inappropriate levity. Compare with a direct version on correct action, scanning, recall, and trust. For translation and regional delivery, test whether the source domain exists and whether the localized expression preserves the mechanism. Relevant communities and domain reviewers interpret harm signals; an unrelated majority vote does not cancel them.

Where it stops holding

Conventional interface models such as “folder” and “shopping cart” may be clearer than a newly invented literal label, and domain metaphors can connect to external knowledge. A metaphor does not fail merely because it has an etymology. Conversely, a replacement is not inclusive merely because it is new; its comprehension and role implications still need testing. Creative marketing can use richer rhetoric, while safety, payment, permission, error, and irreversible actions retain direct statements. Representation and authorization of people imagery belong to visual-asset governance; this leaf evaluates verbal metaphors and their task effects.

Applying it

• Inventory metaphors in high-exposure, high-risk, and cross-region copy. Record source domain, target concept, required background, negative or subordinate roles, plausible harm, task consequence, and owner.
• Write a direct alternative for each consequential metaphor: what object changes, how, and what the user can do. If the direct version suffices, decide from evidence whether the metaphor aids, replaces, or should be removed.
• Register the audience, channel, and context of retained metaphors in terminology and style rules. Let target languages use a different direct expression rather than forcing the source metaphor.
• When migrating established terms, update UI, docs, search, and support together and preserve justified search aliases. Monitor misunderstanding, contacts, and harm reports after launch instead of relying only on a static sensitive-word scan.

Related

• Same group: T1.08.1 Avoid default assumptions about ability, gender, or age · T1.08.3 Self-described identity outranks system classification
• Adjacent: S3.02.3 Physical-object metaphors may lack a local referent · T1.04.3 Term changes require global replacement and migration notes
• Search terms: exclusionary metaphor · direct language · metaphor comprehension