metaphor theoriesOverview

Metaphors are not simply literary devices, but something active in understanding, perhaps even the very basis of language.

Read this section for arguments that metaphors organize our experience, uniquely express that experience, and create convincing realities. Poetry, which uses them instinctively, is following a scientific truth.


Metaphor commonly means saying one thing while intending another, making implicit comparisons between things linked by a common feature, perhaps even violating semantic rules. {1} Scientists, logicians and lawyers prefer to stress the literal meaning of words, regarding metaphor as picturesque ornament.

But there is the obvious fact that language is built of dead metaphors. As a traditional critic put it: "Every expression that we employ, apart from those that are connected with the most rudimentary objects and actions, is a metaphor, though the original meaning is dulled by constant use." Consider the words of that very sentence: an ´expression´ is something squeezed out; to ´employ´ something is to wind it in (implicare ); to ´connect´ is to tie together (conectere); ´rudimentary´ comes from the root to root or sprout; an ´object´ is something thrown in the way; an´action´something driven or conducted; ´original´ means rising up like a spring or heavenly body; ´constant´ is standing firm. ´Metaphor´ itself is a metaphor, meaning the carrying across of a term or expression from its normal usage to another." {2}

Metaphors are therefore active in understanding. We use metaphors to group areas of experience (life is a journey), to orientate ourselves (my consciousness was raised), to convey expression through the senses (his eyes were glued to the screen), to describe learning (it had a germ of truth in it), etc. Even ideas are commonly pictured as objects (the idea had been around for a while), as containers (I didn't get anything out of that ) or as things to be transferred (he got the idea across). {3}

Metaphors in Science

How does science and scientific prose deal with this most obvious of facts? By stratagem and evasion. The scientific style aims at clarity, objectivity and impersonality — attempting to persuade us that the reality depicted is independent of experimenter and reporting. The key evidence is that laid out in the scientific paper, which, though purporting to be a plain account of what was done and observed, is in fact {4} a carefully tailored document making a bid for personal recognition. The abstract allows the significance of the work to be modestly hinted at. The passive voice makes appear inevitable and impersonal what was often achieved only after great effort and skill by the experimenter. Stylistic devices like metaphor, irony, analogy and hyperbole that might call attention to the staged nature of the reporting are muted or banned. Where descriptive, the language employs figures drawn from physics: inert and mechanical. Sentence structure is simple, not to say barbaric: commonplace verbs linking heavy noun clusters. References pay homage to previous workers in the field, and imply familiarity with procedures and therefore professional competence.

Linguistic Philosophy

Can metaphors be paraphrased in literal terms? Many philosophical schools supposed they could, perhaps even needed to do so, particularly those of the Logical Positivist approach who stressed the rational, objective aspects of language. But influential papers by Max Black showed that readers come to metaphors armed with commonplace understandings of the word employed, understandings which enter into how we read the passage. In When sorrows come, they come not in single spies, but in battalions both spies and battalions have different connotations that interact and shape our understanding in ways that escape a literal paraphrase. {5}

Not everyone agrees. As would be expected from a theorist who needs a logically transparent language, Davidson denies that metaphors have a meaning over and above their literal meaning. They may point to some resemblance between apparently dissimilar things, but they don't assert that resemblance and do not constitute meaning. {6}

Lakoff and Johnson

Metaphors are much more powerful instruments in the eyes of Lakoff and Johnson. {7} Metaphors have entailments that organize our experience, uniquely express that experience, and create necessary realities. Lakoff and Johnson attacked the two commonly accepted theories of metaphor. The abstraction theory — that there exists one neutral and abstract concept that underlies both the literal and metaphoric use of word — failed on six counts. The abstraction doesn't apply throughout, in height, emotion, future, etc. We can say A is B, but the reverse, B is A, is not equivalent. The theory doesn't account for the structuring of different aspects of a concept, nor with the fact that when we say A is B, the B is always the more concrete and clearly defined. The systematic way in which metaphors apply is not explained, nor how metaphors are made to fit the occasion. Equally, on several counts, the homonymy theory — that the same word may be used for different concepts — also fails. In its strong form the theory cannot account for relationships in systems of metaphors, nor for extensions of such metaphors. In its weak form the theory doesn't account for categories of metaphor. In addition to the above-mentioned difficulty that B is always more concrete and clearly-defined than A, it is to be doubted that statements like "I'm on a high" really involved similarities at all.

Previous theories derive, Lakoff and Johnson believe, from a naive realism that there is an objective world, independent of ourselves, to which words apply with fixed meanings. But the answer is not to swing to the opposite and embrace a wholly subjectivists view that the personal, interior world is the only reality. Metaphors, for Lakoff and Johnson, are primarily matters of thought and action, only derivatively of language. Metaphors are culturally-based, and define what those with certain assumptions and presuppositions find real. The "isolated similarities" are indeed those created by metaphor, which simply create a partial understanding of one kind of experience in terms of another kind of experience. They are grounded in correlations within our experience.

Evidence of Psychology

If metaphor permeates all discourse, it necessarily played a large part in the history of psychology, particularly in generating fruitful ideas. But metaphor does not simply express, it conditions thought. Psychologists at the turn of the century (and Freudians even today) tended to picture psychic energy as steam in a pressure-boiler. Mind is subsidiary, something brought to life by the energy of the instincts. {8} Deviant behaviour has also been seen as spirit-possession, a pathology, dementia, hallucination, inappropriate response, mental imbalance, spiritual and intrapersonal poverty — views which have not only coloured society's views of the "afflicted" but also guided treatment. {9}

The process continues. Neurological discourse employs metaphors from telecommunications, computer science and control systems. Analysis of emotions revolves round metaphors of inner feelings, driving forces, animal instincts, etc. Motivation looks to metaphors of vigilance and defence. Perception oscillates between mirrors of reflection and moulders of experience. Social analysis uses the concepts of laboratory work, mechanical regulation, meaningful relationships and systems theory. What is the "correct" view? There isn't one. Yet metaphor is not an empty play of words, or even free play of ideas. Metaphors need to be in harmony with the social and historical setting, with the beliefs and personal constructs of the society or micro-society of the time. {10}

Sociology and Anthropology

Sociologist and anthopologists are much interested in metaphor — because of the Whorf-Sapir hypothesis, the supposedly primitive thinking of some tribes, and the clash of cultural contexts implicit in translation. Equally important is the light thrown by the study of native people on our own western cultures and unexamined assumptions. Sociologists remember what Vico said long ago: "man, not understanding, makes his world." Much of man's reasoning is vacuous, simply transferring meaning from intimate, domestic surroundings to the unknown. {11} In less picturesque terms, metaphor is a mapping from source (familiar, everyday) to target domain (abstract, conceptual, internal, etc.) But, contrary to Lakoff and Johnson's view that metaphor represents something fundamental to brain functioning, many sociologists regard the target domain as culturally determined. In describing their marriages, speakers choose models (target domains) that provide a helpful match ("we made a good team, I'd be lost without her"). {12}

How do sentences in different languages have the same meaning? Rationalists assume that there is a universal base of shared semantic primitives (just as Chomsky's grammar once supposed there were syntactic universals) but fail to explain how this base came about. Empiricists argue for some body of shared experience that arises from contact with the real natural world, but can't explain why language takes the form it does. Linguists like Jakendorf suppose that language grows out of perceptual structures — meaning is part of the meaningfulness of experience — but then need to forge detailed links between the two. {13} Jardine believes that all objects are intentional objects — i.e. that our human senses and intelligence are conditional, and restrained by our biologic make-up. Words become components of experience. {14} Alverson {15} considers the preposition "over" from Lakoff's perspective {16} and accepts that schemas are not reducible to propositions, are the core-meanings of words, enter into syntax, are ideal in origin and partly predictive, enter into networks with other schemas, and enter into metaphorical and conventional extensions. But they are not brain-based as such or primitive. Languages contain codings of universal schemas, but their partitioning into words varies with cultural context. Schemas remained as symbols for categories of sense as intention-and-significance-bestowing devices, not abstract configurations.

Literary Use

For writers and critics, metaphor is simply a trope: a literary device deriving from the schools of classical rhetoric and intending to put an argument clearly and persuasively. Boundaries are not sharp, but devices are commonly grouped as schemes and tropes. Schemes, which include alliteration, chiasmus, etc., have more to do with expression. Tropes, which include metaphor, metonymy and synecdoche, are more powerful and deal with content. {17} Metonymy entails using a name to stand for the larger whole: "Whitehall intended otherwise." where Whitehall stands for the British civil service. Metonymy does not open new paths like metaphor, but shortens distance to intuition of things already known. {18} Metaphor therefor involves a transfer of sense, and metonymy a transfer of reference. {19}

There are larger considerations. Kenneth Burke thought tropes were ready-made for rhetoricians because they describe the specific patterns of human behaviour that surface in art and social life. {20} Hayden White sketched a theory of history which bridged the claims of art and science by defining the deep structures of historical thought in terms of metaphor, metonymy, synecdoche and irony. {21} For Derrida, the inevitable clash of metaphors in all writing shows only too well that language may subvert or exceed an author's intended meaning. Like Derrida, Paul de Man saw language as an endless chain of words, which cannot be closed off to a definitive meaning or reference. The literal and figurative meaning of a text is not easily separated, and the realities posited by language are largely those accepted by the dominant ideology as truthful representations of the world. {22}

Rhetoric of Science

Alan Gross goes a good deal further than most literary critics in his Rhetoric of Science. {23} Truth in science, he argues, is a consensus of utterances rather than a fit with evidence. Whatever scientists may assert — and they very much resent any reduction of science to a form of persuasion — philosophers have long known that the claim of science to truth and objectivity rests on shaky foundations. Knowledge does not exist independently of conceptual schemes, or even perhaps of linguistic formulation. Indeed, has not the contemporary logician, W.V.O. Quine, shown that science is under-determined by experience: the edges may square with experience but the interior cannot be more than a coherence view of truth? Perhaps it comes down to practicalities. The sheer bulk of "scientific findings", its dependence on certain procedures and assumptions, not to speak of vetting and reviewing procedures, all of which ensure that the reality which science portrays exists as statements which are now too costly to modify. Of course science "works", but then so does mathematics, which has largely relinquished claims to logical foundations or transcendent truth. {24}


What are Lakoff and Johnson saying but that there is no one central interpretation? Use different turns of speech — as we do naturally in our everyday lives — and the "meaning" alters. Without thinking twice we translate from one mind-set to another. We have probably always done so. {25} Speech started as a primary function in oral societies. There was no "content" behind the expressions. Hieroglyphics were not word pictures but mnemonic devices initially, becoming logograms in Egypt and Mesopotamia in third millennium BC, and only later denoting a syllable sound. It was the North Semitic Byblos alphabet of BC 1400 that the ancient Greeks adapted, turning four of the consonants into vowels that allowed entire speech to be placed on the page, when the focus passed from words to invisible ideas.

What of the Iliad and Odyssey? Parry and other scholars showed that Homer's productions were improvisations to music of a vast collection of stock phrases — a procedure still used by Serbian Guslars who can improvise tens of thousands of lines in this way. Plato preferred the new written procedures (castigating poets of the old oral tradition in The Republic) but also worried that the very process of writing and learning from texts imprisoned speculation in authoritative interpretations. Meditation was needed to bring the past into the presence, and this may also explain Plato's desire for eternal forms. Classical rhetoricians developed mnemonic devices but it was the north European scholastics who made memory a record of doings that could be examined under confession. In twelfth-thirteenth century Europe the validity of an oath (given word, symbolically the Word of God) is transferred to documents that have legal force.

Translation was not an issue in the classical world: the literate spoke several languages and could interpret ( i.e. recast) from one to another. The Christian Church became monolingual to incorporate Greek and Hebrew into the culture of late Antiquity. Indeed, for long centuries, the vernacular spoken by all classes in Europe was a romance language pronounced differently in different places, none of the pronunciations being close to classical Latin. It was never written down, and only in ninth century Germany was an attempt made to create a 'German grammar'. Charlemagne accepted a uniform pronunciation of official Latin, but this was incomprehensible to his subjects and was therefore repealed. Depositions were taken from the vernacular and written in Latin, and Latin creeds were rendered and remembered in the vernacular. Elio Anonio de Nebrija attempted in 1492 to create a Spanish that was not spoken but served to record speech, his grammar and argument for a standardized Castillano being intended to curb the publication of literature inimical to the crown.

Until comparatively recently — continue Illich and Sanders {26} — there was no self as such, but only an "I" that glowed into life as it recounted its adventures or told its autobiography. Chaucer claimed a fantastic memory to avoid the Church's injunction against invention, employing also a complex syntax so that listeners were compelled to imagine the page. The first novel to "make up facts" was Defoe's Journal of the Plague Year, which undercut the dependence on written testimony to which the work alluded. The work was fiction dressed up as fact, just as Huckleberry Finn asks the reader to believe in The Adventures of Tom Sawyer by an illiterate Tom. But his misspellings and incorrect expressions do all the same evoke the great openness and freedom of the meandering Mississippi River, which implies that we are imprisoned by our own mannered language. Coming to modern times, we note that Orwell's Newspeak served as a mechanical substitute for thought, and was therefore a parody of the "Basic English" promulgated in the thirties. And today of course we have the impersonal language of science and business.

Concluding Thoughts

Where is metaphor grounded? Not in logic, nor literary theory. There is no purely literal language in terms of which metaphor may be evaluated and objectively assessed. Along a broad front in cognitive psychology and social anthropology, metaphor is currently subject to extensive analysis, but the findings can only be partial, and relative to the discipline involved. What is becoming clearer is that metaphor — like linguistic theory and theories of speech acts — is rooted in the beliefs, practices and intentions of language users.

This and other pages in the theory section have been collected into a free pdf ebook entitled 'A Background to Literary Theory'. Click here for the download page.


1. Robert R. Hoffman et al.'s Cognitive metaphors in experimental psychology in David Leary's Metaphors in the History of Psychology (1990). Book has an extensive bibliography.
2. p. 193 in F.L. Lucas's Style (1955). Also see J.D. Becker's The Phrasal Lexicon (1975) and bibliography of Hoffman et al.
3. pp.178-179 in Hoffman et al 1990.
4. Alan Gross's The Rhetoric of Science (1990).
5. See Metaphor entry in Ted Honderich's The Oxford Companion to Philosophy and Max Black's More about Metaphor in Andrew Ortony's (Ed.) Metaphor and Thought (1979).
6. Donald Davidson's What Metaphors Mean in Inquiries into Truth and Interpretation (1984).
7. G. Lakoff and M. Johnson's Metaphors We Live By (1986). Andrew Goatly's The Language of Metaphors (1997) is a systematic elaboration.
8. Kurt Danziger's Generative metaphor and the history of psychological discourse in Leary et al. 1990.
9. Theodore Sarbin's Metaphors of unwanted conduct: a historical sketch in Leary et al. 1990.
10. David Leary's Metaphor, theory and practice in the history of psychology in Leary et al. 1990.
11. p. 4 in James Fernandez's Beyond Metaphor: The Theory of Tropes in Anthropology (1991).
12. Naomi Quinn's The Cultural Basis of Metaphor in Fernandez 1991.
13. Ray Jackendorf's Semantics and Cognition (1983).
14. Nick Jardine's The Possibility of Absolutism in D.H. Mellor's (Ed.) Science, Belief and Behaviour: Essays in Honour of R.B. Braithwaite (1980).
15. Hoyt Alverson's Metaphor and Experience: looking Over the Notion of Image Schema in Fernandez 1991.
16. George Lakoff's Women, Fire and Dangerous Things: What Categories Reveal about the Mind (1987).
17. pp 74- 76 in Geoffrey Leech's A Linguistic Guide to English Poetry (1969).
18. p. 153 in Leech 1969.
19. Michael Issacharoff's Jakobson, Roman in Michael Groden and Martin Kreiswirth's (Eds.) The Johns Hopkins Guide to Literary Theory and Criticism (1994).
20. Paul Jay's Burke, Kenneth in Groden and Kreiswirth 1994.
21. Hans Kellner's White, Hayden in Groden and Kreiswirth 1994.
22. Cynthia Chase's Man, Paul de in Groden and Kreiswirth 1994.
23. Gross 1997.
24. See George Lakoff and Raphael Núñez's Where Mathematics Comes From: How the Embodied Mind Brings Mathematics into Being (2000). for an application of metaphor theory to mathematics.
25. I. Illich and B. Sanders's The Alphabetization of the Popular Mind (1989).
26. Ibid.

Internet Resources

1. Understanding the basics of metaphor in poetry . Garry Smith. 2002. Straightforward account of literary use: Dylan Thomas poem example.
2. Glossary of Poetic Terms. Useful definitions, examples and quotations: includes metaphor.
3. Links to Rhetorical Theory Notes. Excellent notes on and reading lists for rhetorical theory.
4. Metaphor and Meaning. William Grey. 2000. Literary use of metaphor in some depth.
5. Kenneth Burke. James F. Klumpp. 2002. Seminar notes.
6. Philosophy and Rhetoric, Argument and Exploration. Doug Brent. Oakeshott and Burke's views of rhetoric.
7. An Irenic Idea about Metaphor. William G. Lycan. Searle and Davidson's arguments in more detail.
8. George Lakoff. Jan. 2004. Introduction to Lakoff and controversies raised.
9. "Metaphors We Live By" by George Lakoff and Mark Johnson. Janice E. Patten. 2003. Review/summary of first four chapters of the book.
10. Cognitive Linguistics and the Marxist approach to ideology. Peter E Jones. Cognitive linguistics and a Marxist critique of ideologies.
11. Does Cognitive Linguistics live up to its name? Bert Peeters. Review of current work in cognitive linguistics.
12. George Lakoff: The Theory of Cognitive Models. Francis F. Steen. Apr. 1997. Critical review of Lakoff's work.
13. Thinking About Thought. Piero Scaruffi. 2001. General essay that takes metaphor theory a little further.
14. Hermeneutics of Metaphor. A theological perspective.
15. Integration and Conceptual Modeling. Thomas J Wheeler. Metaphor and brain physiology
16. Mark Turner. Home site, with publications, etc. and links.