One of the most influential studies of language in this direction was The Meaning of Meaning (1923) by C. K. Ogden and I. A. Richards. Like Owen [Barfield] and [Ezra] Pound, Ogden and Richards see World War I as an extreme example of the tyranny of language and concern themselves with building a theory of language that will get rid of traditional philosophic and religious assumptions about it. They hold to a low evaluation of language. For students of C. S. Lewis, The Meaning of Meaning is usefully considered in conjunction with Owen Barfield's Poetic Diction (1926), which espouses a high evaluation of language. As Barfield explains in the preface to the second edition (1952), he did not originally intend his book to be an answer to Ogden and Richards (15). Nevertheless, the contrast between the two books provides an insight into the intellectual context of Lewis' fiction, for in it Lewis defends a view similar to Barfield's and reacts against the view of Ogden and Richards.1

Like others who hold to a low evaluation of language, Ogden and Richards believe that mankind has always been hampered by language superstition, which they define as the erroneous view "that words . . . always imply things corresponding to them" (31). The seemingly sophisticated philosophy of the ancient Greeks is based on their misconception that language has some necessary relationship to the structure of reality. Out of this misconception they created "the World of Being," in which bogus entities reside" (32). Although language superstition has always been a problem, Twentieth-Century advances in the technology of communication have made it more dangerous by disseminating more linguistic confusion (29).

In order to combat language superstition and get rid of the bogus entities, Ogden and Richards formulate a theory of meanings in terms of Watson's behaviorism (Wolf 86-87). An external object causes a sensation--the modification of a sense organ; repeated encounters with the object will produce similar sensations, and gradually the connection between "symbol" (word) and "referent" (object) will be established by a process similar to the one by which the dinner bell came to have meaning for Pavlov's dogs (Ogden and Richards 53, 56-57).2 This behaviorist formulation eliminates "the primitive idea that Words and Things are related by some magic bond" (47), which leads to the use of symbols that have no referent. The empty use of language leads to the creation of bogus entities and what Ogden and Richards refer to as psittacism (the parrot disease), the inability to free oneself from catchphrases (217-18).

Their theory of metaphor is an extension of this behaviorist formulation. They assert that a metaphor arises when a speaker abstracts similarities between something physical and some other thing. An alternate definition is that metaphor occurs when the properties of a referent within one universe of discourse are applied to another universe of discourse.3 Both of these definitions imply that metaphor is an especially complex kind of abstraction. Indeed, Ogden and Richards state that educated people use metaphor easily while "very simple folk" have small vocabularies based on concrete experience and do not use metaphors" (213-14).4 In a chapter entitled "The Canons of Symbolism" they set up rules for classifying referents and making sure that each symbol stands for only one referent. Since metaphor by definition refers to more than one thing, it belongs only to poetic, emotional, and nonreferential discourse.

This description of how language is or should be used seems counterintuitive, but with it Ogden and Richards propose to exhibit the sources of linguistic confusion and free mankind from language-based irrationality. They do not explicitly state that their method will replace religion, but they do occasionally use salvation rhetoric, as in the assertion that their approach will "free" us from metaphysicians and bishops and "restore our faith" in physicists (83-84).

Busy with the studies that won honors degrees in philosophy, classics, and English, Lewis may not have read The Meaning of Meaning at the time of its first publication in 1923. Indeed, his postwar disillusionment was declining by then. As he recalls in Surprised by Joy, when he returned to Oxford in 1919 he worked to become "a realist," assuming what he calls his "New Look." His realism was chiefly characterized by the determination "never . . . to be taken in again" (204). It probably did not involve an explicit abandonment of the high evaluation of language inherent in his classical education; indeed Ogden and Richards would have regarded his Hegelianism as riddled with bogus entities. Lewis' shaky hold on his New Look was threatened when Owen Barfield, his closest friend, became involved in Rudolf Steiner's anthroposophy. As Lewis remarks in Surprised by Joy, "here [in anthroposophy] was everything which the New Look had been designed to exclude." (206). He and Barfield began to engage in an "incessant discussion" of philosophy, "sometimes by letter and sometimes face to face," which they called "the Great War" (207). Lionel Adey, who has examine the surviving documents of "the Great War" in detail, says that Lewis especially objected to the idea that truth could be grasped by the imagination, to the blurring of the distinction between the real and the imagined, and to Steiner's system of exercises for training one's intuition (Adey 30-31, 51).

Although Lewis could never follow Barfield into anthroposophy, "the Great War" was influential in forcing him to "take that look off [his] face" (Surprised by Joy 217)--that is, to give up his hard-edged realism. The chronology suggests the extent of the influence. According to Adey, Barfield first became interested in Rudolf Steiner and anthroposophy in 1923 (12-13); meanwhile, Lewis' temporary appointment as a philosophy tutor in 1923-24 forced him to reevaluate his "watered Hegelianism" for tutorial purposes (Surprised by Joy 222). Barfield began drafting Poetic Diction in 1924-25, and the bulk of the "Great War" letters were written in 1925-27 (Adey 12-13). As can be seen in his mature writings, Lewis ultimately adopted most of Barfield's theory of language while rejecting its basis in anthroposophy. In Surprised by Joy he says, "Much of the thought which he [Barfield] put into Poetic Diction had already become mine before that important book appeared" (200).

Barfield's whole theory of language is opposed to that of Ogden and Richards.5 Instead of viewing the human being as a passive recipient of sensory stimuli, he sees the mind as an active participant in the very nature of the universe. And instead of regarding metaphor as an abstraction, something added on to more precise, more basic expressions, he regards it as the source of both language and knowledge.

As an anthroposophist, Barfield believes that there is a cosmic Intelligence which is gradually becoming visibly incarnate in human intelligence. A person who is thinking is participating in this cosmic Intelligence. Raw sensory data do not constitute knowledge, or even mental activity; "the pure sense-datum" (48) is merely the percept, meaningless in itself. In order for cognition to occur, the percept must be synthesized with the concept, which Barfield defines as "what I bring to the sense-datum from within" (55). The act of cognition, the synthesis of percept and concept, "creates" the world we experience. His formulation is, of course, reminiscent of Coleridge's primary and secondary imagination, Shelley's understanding of metaphor, the philosophy of the American transcendentalists, and ultimately Plato.

It is not surprising, then, that instead of beginning with behaviorism or any other psychology, he begins with six texts which exemplify poetic language. His first example, a description of a steamship in Pidgin English, is used to show the relationship between language and knowledge: "Thlee-piecee bamboo, two piecee puff-puff, walk-along side, no-can-see" (43). For the South Sea Islander, who speaks pidgin, the words are not poetic. They simply express his concept of the steamship in the words available to him. But for the Englishman, who holds a different concept of the steamship, these words act as poetry. By experiencing the Islander's concept as expressed in pidgin, the Englishman sees the steamship "in a new and strange light" and broadens his understanding (48-49). The words bring him to a new state of awareness, give him a "felt change of consciousness" (32). Poetic pleasure arises form the knowledge that one is moving from a previous state of awareness to a new, expanded awareness. Although the pleasure is momentary, the knowledge of the new concept that came through pleasure is permanent (57).

This opening example of Poetic Diction implies an evaluation of language that is very high indeed. For Barfield, language is not just the symbolization of referents, but a source of knowledge in itself. He believes that concepts are inseparable from the language in which they are expressed, so that the Englishman who has added the pidgin description of the steamship to his linguistic repertoire has added to his inventory of knowledge. Barfield sees a variety of verbal expressions--synonyms and stylistic flourishes--as eminently valuable because each one provides a different perspective to the referent. (His first book, History in English Words, is a fascinating exploration of this view.) Whereas Barfield seeks the enlargement of language, Ogden and Richards seek to control it and narrow its scope. For example, on the basis of the theory of definition set forth in The Meaning of Meaning, Ogden reduced the vocabulary of Basic English to a mere 850-1000 words, a number he deemed sufficient to symbolize most referents.6 In the second edition of Poetic Diction (1952), Barfield explicitly criticizes the Ogden-Richards theory, arguing that the distinction between emotive and referential language is metaphorical.

Barfield's position may be stated in this way: in order to know something, a person must recognize it, and to recognize it, he must be able to relate it to other things. Such relationships are concepts, and concepts must be expressed as resemblances and analogies--metaphors. Since Barfield defines knowledge as "the ability to recognize significant resemblances and analogies" (55), it follows that our knowledge of the universe depends on metaphor. And since human intelligence is a participation in the cosmic Intelligence, the knowledge that human that human beings gain through metaphor corresponds with the way the universe really is. (Ogden and Richards, of course, would dismiss all this as the creation of "bogus entities" through the manipulation of empty verbal shells.)

Philologists before Barfield had discussed the metaphorical basis of language, but he regards their account of why language is metaphorical as mistaken. He cites Max Müller's description of how the Latin spiritus originally meant "blowing, or wind"; then was applied to breath as the physical sign of the principle of life and movement in man and animals; and then became metaphorical as it was applied to the soul, the spiritual principle of a human being. Barfield says Müller is looking at the matter backward, projecting "impossibly modern and abstract concepts . . . on the mind of primitive man" (74). What really happened, according to Barfield, was that primitive man had a single term that meant "blowing," "life," and "soul" simultaneously. There was no separation between literal and metaphorical sense of words (75). It is only in relatively modern times that human beings have developed the ability to create abstractions and therefore split apart the complex cluster of meaning. We have some unsplit words even today. Barfield's example is stomach: when we say, "I have no stomach for such cruelty," we are not using a mere metaphor for a psychological, nonmaterial condition; we are referring to the mental and physical conditions simultaneously or, more accurately, to a single condition that cannot be classified as exclusively one or the other (80n).

Barfield denies the validity of the Ogden-Richards dictum that clear thinking depends on the accurate use of nonmetaphorical, referential language--that is, language that refers to objects, preferably things that modify the sensory receptors. People who hold this belief, Barfield says, do not realize that scientific terms are just as figurative in origin as any other words. He adds, "Those who profess to eschew figurative expressions are really confining themselves to one very old kind of figure" (134). He cites several etymologies to prove that there is no such thing as perfectly literal language, because all the thousands of abstract terms and nuances of meaning were originally references either to objects or to human or animal activity. He calls such terms "fossilized metaphor," observing that "an apparently objective scientific term like elasticity, . . . and the 'metaphysical' abstract . . . are both traceable to verbs meaning 'draw' or 'drag'" (64).

Barfield also denies the validity of separating language from poetic language and ascribing truth only to the former: "The fashionable distinction between Poetry and Science as modes of experience [is] essentially parochial" (138). The revolutionary changes in scientific thought are shifts in consciousness--poetic, metaphorical insights--that apply new concepts to the phenomena of the physical world. Bacon's insight in applying the term "mechanical" to natural principles, Newton's shift of the term "gravity: away from the simple meaning of "weight," Kepler's application of "focus" (etymologically "hearth") to geometry--all these are examples of "the poetic diction of science" (137-38). These scientists increased the store of human knowledge by giving us new metaphors. The much-touted scientific method, says Barfield, is not a way of knowing, but rather a way of testing what has been discovered by insight into metaphorical language (139).7

The unity of metaphor and concept, of language and experience, of poetry and science, is possible, Barfield says, because language is related to nature. Language is metaphorical and mythical because it reflects the true character of the universe. He refers to Emerson's view that such relationships as dark-evil, winter-old age, and breath-spirit are "constant, and pervade nature" (92). Barfield's view leads to the recognition of what we now call Jungian archetypes, although he did not know Jung's work at the time.8

The force of Barfield's view of metaphor, archetype, and myth may be seen by comparing several famous epigrams. First, Max Müller's "Mythology is a disease of language" implies that man's (deluded) belief in divinities is caused by the inevitable distortions in knowledge derived from language. Barfield contradicts Müller, stating that mythological words such as panic, hero, fortune, and fury are derived from primitive man's insights into the original, unified meanings given by nature (89). Ogden and Richards' position is in agreement with Müller's: the mythological words represent a distortion in knowledge derived from a mistaken belief that language has independent power and reality, a distortion that gives rise to empty metaphysical terminology. Their epigram is "Metaphysics is a hybrid of science and poetry," which, like all hybrids, "is sterile" (Meaning 82n). It is a foreshadowing of A. J. Ayer's emendation of Müller: "Metaphysics is a disease of language." In contrast, Barfield, whose theory of language and metaphor asserts the validity of both language and myth, says "Mythology is the ghost of concrete meaning" (Poetic Diction 92).9

The extent to which Lewis adopted Barfield's position is best seen in his essay "Bluspels and Flalansferes"10 composed after The Pilgrim's Regress and first published in Rehabilitations in 1939 and later reprinted in Selected Literary Essays. In it he concludes, with Barfield, that the original metaphors are indeed archetypes--that the universe is indeed organized according to "a kind of psycho-physical parallelism" without which "all our thinking is nonsensical" (265). He refuses assent to Barfield's assertion that one cannot refer to physical objects, or rather one's percepts of them, without using metaphor. But he does conclude that the person who speaks about anything other than a physical object has no choice but to speak metaphorically. Any time we think of causes, relations, or mental states, we inevitably use metaphor. At best we have a choice of metaphors, or the possibility of rapidly shifting from one to another (262-63). Although modern psychologists claim to give us an unmetaphorical account of the soul, their technical terms such as "complexes, repressions, censors, engrams and the like" are metaphorical in origin, so that they are speaking of "tryings-up, shovings-back, Roman magistrates, and scratchings" (261).11 The tough-minded, anti-mystical, "literal" philosophers (like Ogden and Richards) use more meaningless verbiage than Plato, one of "the great creators of metaphor," or the writers of religious literature such as Bunyan and Dante (264, 265).

Lewis also opposes Barfield's formulation by denying that poetry, the exercise of imagination, leads to knowledge. Imagination, he says, is the organ of meaning, not truth. Before a statement can be judged as true or false it must have meaning. Imagination provides the meaning, but reason must determine whether the statement is true or false. He does not, in this relatively brief essay, define what he means by "reason": however, Adey concludes from ""Bluspels and Flalansferes" that Lewis finally did accept Barfield's belief about imagination (82). Perhaps the distinction between meaning and truth is primarily a way to separate himself from Barfield's anthroposophy. It is certain that he makes use of the "psycho-physical parallelisms" of the archetypal metaphor and adds to our verbal repertoire in Barfieldian ways in The Pilgrim's Regress.

Notes

1 Lewis and Barfield met at Oxford, just as Ogden and Richards met at Cambridge. After he wrote Poetic Diction and History in English Words (1926), financial need forced Barfield to leave Oxford and become a solicitor. Ogden is best known for his invention of Basic English, fictionalized as Newspeak in George Orwell's Nineteen Eighty-Four. He translated Wittgenstein's Tractatus-Logico Philosophicus in 1922 and was well acquainted with Russell's thought (Wolf 86-87, 93-94). I . A. Richards developed the low view of language into the New Criticism, sometimes called Formalism, that dominated English studies during Lewis' career.
2 They use the term "engram" (literally "scratching"), coined by Semon, to explain how this comes about. Physical stimuli leave "residual traces" on the brain that contribute to the individual's present response to a similar stimulus. See Wolf's critique of the engram theory, 97-99.
3 Here they anticipate the Chomskyan definition of metaphor in terms of the violation of selectional restrictions.
4 One can only speculate that they had never conversed with country folk, or even Cockneys.
5 The first edition of Poetic Diction implicitly refutes The Meaning of Meaning, although Barfield notices it only with a passing remark that the authors have written "a long and clever book on Meaning" without grasping its essential characteristic, the relationship of meaning to metaphor (134). In the preface to the second edition, he asserts that his theory of poetry is a refutation of The Meaning of Meaning and linguistic analysis in general--that it is, in fact, "not merely a theory of poetic diction, but a theory of poetry: and not merely a theory of poetry, but a theory of knowledge" (14). Although I used all three editions in the preparation of this manuscript, all page references are to the third edition.
6 Basic English is the model for Orwell's Newspeak in Nineteen Eighty-Four, the basic premise of which is that the reduction of vocabulary leads to the destruction of thought.
7 Barfield's insight is a remarkable anticipation of Thomas Kuhn's "paradigm shift" in The Structure of Scientific Revolutions.
8 See the Afterword (1972) to Barfield's Poetic Diction, 218, 221.
9 Barfield's epigram harmonizes with J. R. R. Tolkien's tart comment in 1938 that mythology is no more a disease of language than thinking is a disease of the mind. Incidentally, Tolkien adds another epigram to the nosegay: "The incarnate mind, the tongue and the tale are in our world coeval" ("On Fairy-Stories" 21-22).
10 The unspellable title of the essay is suggested by a passage in Poetic Diction in which Barfield imagines what would happen if Shelley's line "My soul is an enchanted boat" in Prometheus Unbound should give rise to a new word, "chambote." The original meaning of the word would be "the concept 'soul' as enriched by Shelley's imagination" (66), and it would exist in accordance with B's assertion that our concepts of this world are "reflected" in words. Barfield then goes on to speculate about what might happen as the metaphorical nature of the original words was forgotten. In Studies in Words (1960), Lewis returned to "chambote" to spell our exactly what Shelley's imagination would add to the concept "soul" (217). Barfield's complain (in 1985) that scholars are inaccurate when they attribute "to Lewis ideas or opinions they find in Barfield" (C. S. Lewis 111) must be balanced against this verbal resemblance.

It is interesting to note that Ogden and Richards create a word, "wousin," to stand for "the group of phenomena involved or connected in the transit of a negro over a rail-fence with a melon under his arm while the moon is just passing behind a cloud" and use it to make the point that there is no necessary relationship between the symbol and its referent, that "the primitive idea that Words and Things are related by some magic bond" is false (46-47).
11 Between the publication of The Meaning of Meaning in 1923 and Lewis' essay in 1939, others had augmented the theory of Ogden and Richards by asserting that even though such words are metaphorical in origin, they are now dead as metaphor. Composition textbooks began to include exercises of classifying metaphors as Living, Dead, or Dying (no longer felt as figurative but not yet arbitrary). Students were told that the use of dying metaphors would lead to solecisms and mental confusion and were marked down for them.