Mixed metaphors and catachresis as signs of reflexive figurative language
Parent: Writing and Documentation · Published reference · snapshot 2026-09-23 · skill writing-expert/references/mixed-metaphors-and-catachresis.md
↓ Facts as markdownall context files
Source-anchored research for Mixed metaphors and catachresis as signs of reflexive figurative language. Run status: COMPLETED.
These notes link each claim to its source. Some citations identify archived research reports that are no longer hosted on this site. A published reference means the content is available; it does not certify independent review or accuracy.Read the editorial policy and follow the sources before relying on a claim.
Structure and components
- Lakoff holds that conventional ('dead') metaphors are used constantly and automatically, without effort or awareness, yet still structure thought; his 1986 paper title 'The Death of Dead Metaphor' rejects the inert view. (medium confidence) [source]
How it works
- Career-of-metaphor theory: novel metaphors are processed by comparison, and with repeated use the base term gains a domain-general category sense, so processing shifts to categorization and gets faster. This gives a mechanism for deadness as a gradient. (medium confidence) [source]
- Because deadness is relative to reader and context, extending a dead engineering metaphor literally (for example, a bug that 'crawls') can revive its imagery; this is the point where dead metaphors start to collide. (low confidence) [source]
- MIPVU labels a word metaphor-related when its contextual meaning contrasts with a more basic meaning and can be understood by comparison with it. It is a per-lexical-unit procedure; two colliding domains would be found only by labelling each word separately and then comparing their source domains. (medium confidence) [source]
- Incongruity-based metaphor detection is invalid for conventional metaphors, because the surrounding context is usual for the target word. A collision signal is therefore strongest for novel or freshly chosen figures and weak for dead ones. (medium confidence) [source]
- Quintilian (Institutio Oratoria 8.6.34-35) defines catachresis (abusio) as adapting the nearest available term for something with no proper term, and distinguishes it from metaphor, which is used when another term is available. (medium confidence) [source]
- Catachresis has several senses in current usage: categorical boundary crossing when no term exists (chair 'legs'), sound-alike substitution (ravished for famished), strained extension (Shakespeare's 'deepest winter in Lord Timon's purse'), and vaguer synonym substitution (job-seeker for unemployed). (low confidence) [source]
Measurements and reference values
- Kimmel (Journal of Pragmatics, 2010) analysed a corpus of British newspaper commentaries on the 2004/05 EU referenda (2,574 metaphors) and reports mixed metaphors at 39-62% of metaphors, with almost all ontologically mixed clusters comprehensible; he argues mixing is cognitively successful because it covers different aspects of a target domain. Figures are from search-result abstracts; the paper page returned HTTP 403. (low confidence) [source]
Problems, failure modes and limitations
- Merriam-Webster defines a mixed metaphor as a figure of speech combining inconsistent or incongruous metaphors, and lists it beside catachresis as strained use of words that may occur in error or for rhetorical effect. (low confidence) [source]
- A mix of metaphors is therefore not by itself evidence of error or machine authorship: classical rhetoric permits catachresis where no term exists, and corpus work finds human mixing is common and usually coherent. (low confidence) [source]
Comparisons and alternatives
- Scholars disagree on whether dead metaphors are dead: Max Black treats them as separate vocabulary items rather than metaphors, while R.W. Gibbs argues they retain metaphorical qualities, and Turbayne and Walton argue they keep influencing thought. (medium confidence) [source]
- Reed College's writing lab says a string of three or four metaphors leaves readers confused about which model is accurate, or exhausted by repetition, and advises discussing the topic directly rather than piling on figures. (low confidence) [source]
Facts and statements
- Lakoff defines a conceptual metaphor as a systematic mapping from a source domain (e.g. journeys) to a target domain (e.g. love); the mapping licenses source-domain language and inference patterns for the target. (high confidence) [source]
- Source domains are typically more concrete than targets; ARGUMENT is understood via WAR, so 'attack' and 'win' apply to arguments. (high confidence) [source]
- Software-development prose draws on a small set of source domains: construction (building a house), agriculture (garden), skilled craft, and finance (technical debt with interest). (medium confidence) [source]
- Each source-domain lexical item may or may not use the mapping; conventional words in the source domain are not always conventional in the target, so extending a metaphor with new source vocabulary is a novel act, not automatic. (medium confidence) [source]
- A corpus study on Common Crawl (Aono et al., 2024) found verb metaphors take direct objects with lower concreteness, imageability and familiarity, and appear more in emotional, subjective sentences; this supports abstract-target claims but suggests technical, unemotional prose may use fewer live metaphors. (medium confidence) [source]
- Overextending a source domain beyond the initial need creates false equivalences and limits reflection. (low confidence) [source]
- Critics say conceptual metaphors are often set up from intuitive, unsystematic evidence, and corpus-based validation is limited. (low confidence) [source]
- Orwell distinguishes a dead metaphor, one that has reverted to being an ordinary word, from a dying metaphor, one that has lost evocative power and is used without knowledge of its meaning; he objects to the dying kind, not the dead kind. (medium confidence) [source]
- Software bug and thinking-machine constructs such as if/then, call/return and while are cited as dead metaphors in engineering language; readers no longer picture a cockroach. (low confidence) [source]
- A position paper argues that resurrecting the literal origin of dead technical terms (upper case, leading, gutter, out of sorts from typesetting) restores evocative power and aids memory; it reports no empirical results. (low confidence) [source]
- A copyeditor-oriented guide treats a mixed metaphor as two unrelated images describing one thing (e.g. 'iron out the remaining bottlenecks') and tells editors to help authors spot and untangle it, with no detailed procedure. (low confidence) [source]
- Both fetched sources point toward the same edit: remove the conflicting image and keep a single coherent one, or state the point literally. (medium confidence) [source]
- Research from the VU metaphor lab frames whether a mixed metaphor is a fault as a question of deliberateness, so an editor should check for intent before rewriting. Only the title and search snippet were read; the PDF did not parse. (low confidence) [source]
- Cross-dataset evaluation finds LLM metaphor detection tracks lexical overlap and sentence length, and accuracy on literal paraphrases often drops or paraphrases keep the metaphor. An LLM-based collision detector should not be trusted without a check against a source-domain lexicon. (low confidence) (verified as of 2025-07-21) [source]
- Quintilian (8.6.50) warns that many writers begin a figure with a tempest and end with a fire or a falling house, producing a hideously incongruous effect, and says never to mix metaphors. (low confidence) [source]
Children
- Career of metaphor (frontier)
- Conceptual metaphor source domains in technical prose
- Dead versus live metaphors in engineering writing
- Deliberate metaphor theory (frontier)
- Edit protocol for colliding metaphors
- Kimmel corpus study of mixed metaphor (frontier)
- Metaphor collision as a detection signal
- Metaphor conventionality gradient (frontier)
- Mixed metaphor and catachresis
- Per-clause MIPVU-style word labelling (frontier)
- Quintilian on abusio (frontier)
- Read-aloud literal visualization test (frontier)
- Source-domain lexicon overlap check (frontier)
- Technical debt as finance metaphor (frontier)
- War and journey metaphors in engineering (frontier)
Frontier under this node: Career of metaphor, Deliberate metaphor theory, Kimmel corpus study of mixed metaphor, Metaphor conventionality gradient, Per-clause MIPVU-style word labelling, Quintilian on abusio, Read-aloud literal visualization test, Source-domain lexicon overlap check, Technical debt as finance metaphor, War and journey metaphors in engineering