The Great Code

The Bible and Literature

by Northrop Frye

"The Bible" has traditionally been read as a unity, and has influenced Western imagination as a unity. It exists if only because it has been compelled to exist. Yet, whatever the external reasons, there has to. be some internal basis even for a compulsory existence. Those who do succeed in reading the Bible from beginning to end will discover that at least it has a beginning and an end, and some traces of a total structure. It begins where time begins, with the creation of the world; it ends where time ends, with the Apocalypse, and it surveys human history in between, or the aspect of history it is interested in, under the symbolic names of Adam and Israel. There is also a body of concrete images: city, mountain, river, garden, tree, oil, fountain, bread, wine, bride, sheep, and many others, which recur so often that they clearly indicate some kind of unifying principle. That unifying principle, for a critic, would have to be one of shape rather than meaning; or, more accurately, no book can have a coherent meaning unless there is some coherence in its shape.

I am not concerned with the true meaning of such words as episcopos or ecclesza , but, for the most part, with
nouns so concrete that it is practically impossible for any translator to get them wrong.

In this volume such features as the categories of metaphor, the ladder of "polysemous sense," the conception of literal meaning, and the identification of mythology and literature are presented in what I hope is a new framework.

I am not dispensing with the quality of irony that all teachers from Socrates on have found essential. Not all elusiveness, however, is merely that. Even the parables of Jesus were ainoi, fables with a riddling quality. In other areas, such as Zen Buddhism,
the teacher is often a man who shows his qualifications to teach by refusing to answer questions, or by brushing them off with a paradox. To answer a question (a point we shall return to later in the book) is to consolidate the mental level on which the question is asked. Unless something is kept in reserve, suggesting the possibility of better and fuller questions, the student's mental advance is blocked.

I have spoken of my wish to get clear of conventional aesthetic canons, but "unity" is one of those canons, and the Bible's disregard of unity is quite as impressive as its exhibition of it. Ultimately, as we should expect, the Bible evades all literary criteria. As Kierkegaard said, an apostle is not a genius--not that I ever found "genius" a very useful word either. My experience in secular literature had shown me how the formal principles of literature had been contained within literature, as the formal principles of music, embodied in sonata, fugue, or rondo, have no existence outside music.
But here is a book that has had a continuously fertilizing influence on English literature from Anglo-Saxon writers to poets younger than I, and yet no one would say that the Bible "is" a work of literature. Even Blake, who went much farther than anyone else in his day in identifying religion and human creativity, did not call it that: he said "The Old and New Testaments are the Great Code of Art," a phrase I have used for my title after pondering its implications for many years.

Man lives, not directly or nakedly in nature like the animals, but within a mythological universe, a body of assumptions and beliefs developed from his existential concerns. Most of this is held unconsciously, which means that our imaginations may recognize elements of it, when presented in art or literature, without consciously understanding what it is that we recognize. Practically all that we can see of this body of concern is socially conditioned and culturally inherited Below the cultural inheritance there must be a common psychologica inheritance, otherwise forms of culture and imagination outside our own traditions would not be intelligible to us.

Why does this huge, sprawling, tactless book sit there inscrutably in the middle of our cultural heritage like the "great Boyg" or sphinx in Peer Gynt, frustrating all our efforts to walk around it?

In English literature the canons of criticism were established mainly by Samuel Johnson, who followed the normal Protestant practice of keeping the poetic aspect of the Bible in a separate compartment from secular literature. It was the Romantics who realized that such a separation was irrational. Coleridge's brilliant insights into Biblical typology make it clear that he would have made things much easier for his students, and more productive for his influence, if he had provided an interconnected statement of his views on that subject.

A good deal of the Orient is committed to
Marxism, which is the direct heir of the revolutionary and socially organized forms of religion derived from the Bible. Recently a Chinese student, a teacher in own country and about to return there, asked me how he could explain the cultural importance of Christianity for the West to his students in a way that would be intelligible to them. I suggested that they would have some understanding of Marxism, that Marx's spiritual father was Hegel, and that therefore his spiritual grandfather was Martin Luther.

Information does have to be conveyed in teaching, of course, but for the teacher the imparting of information is again in a context of irony, which means that it often looks like a kind of game. When the subject to be taught is literature, this element of game takes on a special appearance. Literature continues in society the tradition of myth-making, and myth-making has a quality that Levi-Strauss calls bricolage, a putting together of bits and pieces out of whatever comes to hand. Long before Levi-Strauss, T. S. Eliot in an essay on Blake used practically the same image, speaking of Blake's resourceful Robinson Crusoe method of scrambling together a system of thought out of the odds and ends of his reading.

Seven phases of what is traditionally called revelation: creation, exodus, law, wisdom, prophecy, gospel, and apocalypse. Two forms of apocalyptic vision are postulated, making eight in all
, the eighth bringing us back to the central thesis of the role of the reader. Then comes an inductive survey of, first, the imagery, and then the narrative structures of the Bible, which is the point from which the book took its origin. The final chapter makes a second approach to the "rhetoric of religion," and includes a brief sketch ofa "polysemous" or multileveled conception of meaning as applied to the Bible. The latter attempts to suggest some answers to questions about the direction in which we go from the "literal" meaning.

I should hope that a further study would bring us closer to commentary in more detail on the text of the Bible. Ruth, the Song of Songs, and the folktale material in the Apocrypha have a particularly obvious literary reference and yet get very short shrift here.


Language I

Everyone concerned with language is aware of the extent to which reading a translation is a settling for the second best. This is particularly true of major poetry, where translation has to be a miracle of tact, and even then does not claim to replace its original. On the other hand, abstracts of articles in scientific or mathematical journals can easily be translated or even read in the original by those with limited command of the language, because there is a third underlying language of subject matter which is international. The Bible, however, seems much closer to the poetic area than to the scientific-journal area. Clearly, then, one of our first problems is to determine the positive reality of translation, the essential thing or force or process that translation translates.

This question normally starts with a rough-and-ready distinction between sound and sense. The sound-associations within a language cannot usually be translated adequately, although they are of immense importance in building up linguistic responses.
This fact has nothing to do with whether philology recognizes the associations as genuine within its own area. The assonances between words of similar reference (e.g., "God" and "good" in English), the standard rhymes. the words of multiple meanings that allow for puns, are all accidents. or, as philologists like to say, "pure" coincidences; yet they make up a texture that enters into the mental processes of all native speakers of the language, whether they are writers or not. Such a texture. extending as it does to a dense mass of idioms that can often be translated only by a complete rephrasing of the original, helps tmake language one of the most fragmented of all human phenomena.

What can be translated, we assume, is
that particular relation between different signifiers to a common signified that is known as "sense." To use a convenient French distinction: there is, in addition to the langue that separates English and French and German, also a langage, that makes it possible to express similar things in all three languages.

It is not necessary to invoke any more subtle entities, such as Jung's collective unconscious, to explain the fact that human creative expression all over the world has some degree of mutual intelligibility and communicating power. We note further that Luther's German Bible, and the sequence of English Bibles culminating in the AV, were powerful generators of imagery, narrative, allusion, and other forms of verbal articulateness in their cultures; the same could be said of many Classical and other translations. What we call langage, then, is a very positive linguistic force. One wonders whether it is substantial enough for there to be such a thing as a history of langage, a sequence of modes of more or less translatable structures in words, cutting across the variety of langues employed, affected and conditioned but not wholly determined by them. Such a possibility, if it could become anything more than that, would provide a historical context for the Bible of a type that I do not think has yet been examined.

This question took me to Vico, the first person in the modern world to think seriously about such matters. According to Vico, there are
three ages in a cycle of history: a mythical age, or age of gods; a heroic age, or age of an aristocracy; and an age of the people, after which there comes a ricorso or return that starts the whole process over again. Each age produces its own kind of langage, giving us three types of verbal expression that Vico calls, respectively, the poetic, the heroic or noble, and the vulgar, and which I shall call the hieroglyphic, the hieratic, and the demotic. These terms refer primarily to three modes of writing, because Vico believed that men communicated by signs before they could talk. The hieroglyphic phase, for Vico, is a "poetic" use of language; the hieratic phase is mainly allegorical; and the demotic phase is descriptive. Vico's three terms, apart from their identification with writing, are extremely suggestive as providing a starting point for thinking about the place of the Bible in the history of language as langage, though in what finally emerged for me very little of Vico was left. The sequence of literary modes in my Anatomy of Criticism is much closer to Vico, but that relates to a different set of phenomena, as I shall try to show.

I think we can see in most Greek literature before Plato, more especially in Homer, in the pre-Biblical cultures of the Near East, and in much of the Old Testament itself,
a conception of language that is poetic and "hieroglyphic," not in the sense of sign-writing, but in the sense of using words as particular kinds of signs. In this period there is relatively little emphasis on a clear separation of subject and object: the emphasis falls rather on the feeling that subject and object are linked by a common power or energy. Many "primitive" societies have words expressing this common energy of human personality and natural environment, which are untranslatable into our normal categories of thought but are very pervasive in theirs: the best known is the Melanesian word mana. The articulating of words may bring this common power into being; hence a magic develops in which verbal elements, "spell," and "charm," and the like, play a central role. A corollary of this principle is that there may be a potential magic in any use of words. Words in such a context are words of power or dynamic forces.

Thus knowing the name of a god or elemental spirit may give the knower some control over it; puns and popular etymologies involved in the naming of people and places affect the character of whatever thing or person is given the name.
Warriors begin battles with boasts that may be words of power for them: boasting is most objectionable to the gods for a corresponding reason: the possibility of man's acquiring through his words the power that he clearly wants. The vow that cannot be broken, including the rash vows that begin so many folktales, as in Jephthah's "I have opened my mouth unto the Lord, and I cannot go back" ( Judges 11:35), again expresses the sense of quasi-physical power released by the utterance of words. When a sacrosanct myth is read at a religious ritual, as, say, the Babylonian creation myth Enuma elish was read at the New Year, some kind of magical energy is clearly being released. It would perhaps be overconceptualizing to say that it was thought to encourage the natural cycle to keep turning for another year; but where the subject and the object are not clearly separated, and there are forms of energy common to both, a controlled and articulated expression of words may have repercussions in the natural order.

All words in this phase of language are concrete: there are no true verbal abstractions. Onians' monumental study of Homer's vocabulary, Origins of European Thought, shows how intensely physical are such conceptions as soul, mind , time, courage, emotion, or thought in the Homeric poems. They are solidly anchored in physical images connected with bodily processes or with specific objects. Similarly the word kairos, which came to mean a crucial moment in time, originally meant the notch of an arrow. What this means from the critical point of view is that while Homer's conceptions would not have been metaphorical to him (when he uses a figure of speech it is usually a simile), they have to be metaphorical to us. As we think of words, it is only metaphor that can express in language the sense of an energy common to subject and object. The central expression of metaphor is the "god," the being who, as sun-god, war-god, sea-god, or whatever, identifies a form of personality with an aspect of nature.

The operations of the human mind are also controlled by words of power, formulas that become a focus of mental activity. Prose in this phase is discontinuous, a series of gnarled epigrammatic and oracular statements that are not to be argued about but must be accepted and pondered, their power absorbed by a disciple or reader. Pre-Socratic philosophers such as Heraclitus or Pythagoras seem to have been essentially oral teachers or gurus; and what has survived from them consists mainly of discontinuous aphorisms with a cosmological reference, like the "all things flow" of Heraclitus. We shall return to this feature of discontinuity at the end of the book.

With Plato we enter a different phase of language, one that is
"hieratic," partly in the sense of being produced by an intellectual elite. I am speaking here not of ordinary language but of the culturally ascendant language, a language that, at the time or later, is accorded a special authority by its society. In this second phase language is more individualized, and words become primarily the outward expression of inner thoughts or ideas. Subject and object are becoming more consistently separated, and "reflection ," with its overtones of looking into a mirror, moves into the verbal foreground. The intellectual operations of the mind become distinguishable from the emotional operations; hence abstraction becomes possible, and the sense that there are valid and invalid ways of thinking, a sense which is to a degree independent of our feelings, develops into the conception of logic. What Homeric heroes revolve in their bosoms is an inseparable mixture of thought and feeling; what Socrates demonstrates, more especially in his death, is the superior penetration of thought when it is in command of feeling.  The basis of expression here is moving from the metaphorical, with its sense of identity of life or power or energy between man and nature ("this is that"), to a relationship that is rather metonymic ("this is put for that"). Specifically, words are "put for" thoughts, and are the outward expressions of an inner reality. But this reality is not merely "inside." Thoughts indicate the existence of a transcendent order "above," which only thinking can communicate with and which only words can express. Thus metonymic language is, or tends too become analogical language, a verbal imitation of a reality beyond itself that can be conveyed most directly by words.

The basis of Plato's use of language is the teaching method Socrates; and Socrates, unlike his predecessors in Greek philosophy professed not to know anything but only to be looking for something.
His celebrated "irony" was a momentous step in transforming the use of language: it implied renouncing the personal possession of wisdom in favor of an ability to observe it. Wisdom so observed emerges from a dialogue or group discussion, typically the symposium, but, with Socrates usually acting as a guide, it seems to take on a direction and purpose of its own, and eventually enters its real home, a world of ideas, where it can be followed only by the intellectual soul within the body of the seeker. Plato is a very great literary artist, but his greatness has much to do with the break that he made from typically literary forms of expression. The first phase of language, being founded on the metaphor, is inherently, as Vico says, "poetic"; the second phase, which is Plato's, retreats from the poetic into the dialectical, a world of thought separate from and in some respects superior to the physical world of nature.

Socrates does not, like Heraclitus, utter discontinuous aphorisms to be pondered and assimilated, though he quotes one or two from oracles, but
orders his discussion in a sequacious argument. The argument, like the argument of the epic in a different way, starts in the middle and moves both backward and forward: backward to definitions of the terms used, forward to the consequences and implications of adopting these definitions. Eric Havelock, in A Preface to Plato, associates the Platonic revolution in language with the development of writing, which was originally confined mainly to commercial transactions but was now extending itself into culturally ascendant areas. For my purposes, however, it will be more useful to associate the Platonic revolution with the development of continuous prose. Continuous prose, though often regarded, with Moliere's Jourdain, as the language of ordinary speech, is a late and far from "natural" development, and is much less direct and primitive than verse, which invariably precedes it in the history of literature. The language of ordinary speech, as I have tried to show elsewhere, has a loose associative rhythm quite different from actual prose.

Plato's interest in mathematics is consistent with his use of language, for there are obvious
metonymic features in mathematics. In Euclidean geometry, for example, the drawn line, which necessarily has some breadth, is "put for" the ideal or conceptual line that is length without breadth; similarly with the conception of abstract number apart from a number of things. One feels that some of the pre-Socratics and atomic philosophers, such as Anaxagoras or Democ-ritus, were moving more directly from metaphor toward what we should think of as science, from gods to the operations of nature, and that Plato turns away from this direction, toward a transcendent world rather than an objective one. The Timaeus seems to be involved primarily with the degree t(rwhich nature conforms to conceptual models, and in the Phaedo this sense of aesthetic conformity seems to be linked to matters of faith. But what may seem in hindsight to be a retrograde tendency may be less so in the perspective we are trying to attain here

In Raphael's School of Athens Plato points to heaven and Aristotle to the earth, but as far as his main historical influence goes Aristotle points straight ahead. He worked out the organon of a deductive logic based on a theory of multiple causation, and provided
a technique for arranging words to make a conquering march across reality, subjects pursuing objects through all the obstacles of predicates, as the Macedonian phalanxes of his pupil Alexander marched across Asia. But it was a long time before his techniques could really be absorbed by later thinkers. In the later Classical period Plato's sense of a superior order that only language, in both its verbal and mathematical forms, can approach merges with the conception generally identified as logos. This is a conception of a unity of consciousness or suggested by the fact that properly constructed verbal sequences seem to have an inherent power of compelling assent. In Stoicism, and in Christianity in a different way from the beginning, the conception of logos acquires both a religious and a political dimension: it is seen as a possible means of uniting human society both spiritually and temporally.

In metaphorical language the central conception which unifies human thought and imagination is the conception of a plurality of gods, or embodiments of the identity of personality and nature.
In metonymic language this unifying conception becomes a monotheistic "God," a transcendent reality or perfect being that all verbal analogy points to. Such conceptions as the Form of the Good in Plato or the Unmoved Mover in Aristotle are not difficult to absorb into this idea of God, but the Zeus of Homer is more recalcitrant. A monotheism in which one god is supreme over all other gods exists a different linguistic context from a monotheism in which "other gods" do not and cannot exist at all, at least as fully divine beings. In Homer, however, there is sometimes the suggestion that Zeus is not merely the king of gods but contains all the other gods, as in the passage in the Iliad (viii) where he tells the squabbling subordinate deities that he holds heaven and earth, including them, on a gigantic chain that he can at any time pull up into himself. This form of metaphor, which unites the group and the individual, will be of great importance in our argument later, and the passage is also a portent of the great metonymic conception of the chain of being, of which also more later. In any case the word "God," however great its number of referents, is practically a linguistic requisite for metonymic thinking. There is no point in making analogical constructs out of words unless we have something to relate the analogy to.

As Christian theology gained cultural ascendancy, thought began to take on a deductive shape in which everything followed from the perfection of God, because of the need for irrefutable premises. In this process certain tensions were bound to arise with the more metaphorical constructs of earlier ages, when metonymic thinkers were compelled to take them seriously. The tension expresses itself in a moralizing and rationalizing approach to them: if God says or does A , then he cannot also say or do B , if B is inconsistent with A . There are some tendencies in this direction within the Bible itself: compare, for example, II Samuel 24:1 and I Chronicles 21:1. Paganism had similar difficulties, and the metaphorical element in "indecent" or morally paradoxical stories about the gods, found in Homer and elsewhere, had to be deconstructed and assimilated to other linguistic procedures. This was normally done through allegory, which is a special form of analogy, a technique of paralleling metaphorical with conceptual language in which the latter has the primary authority. Allegory smooths out the discrepancies in a metaphorical structure by making it conform to a conceptual standard.

What makes this possible is the development of continuous prose, the main instrument of thought in the metonymic period. In continuous prose, if A and B seem to be inconsistent, one can always insert intermediary verbal formulas, or rephrase them in a commentary, in a way that will "reconcile"
them: if only we write enough of such intermediate sentences, any statement whatever can eventually be reconciled with any other statement. Commentary thus becomes one of the leading metonymic genres, and the traditional metaphorical images are used as illustrations of a conceptual argument.

In Christian theology the principle of analogy can readily be invoked without recourse to allegory. In the Summa contra Gentiles (I, 96) we read "That God hates nothing." In St. Thomas's metonymic context, such a proposition is practically self-evident: no perfect being could hate anyone or anything without ceasing to be a perfect being. Faced with the list of things in the far more metaphorical Bible that God is explicitly said to hate, St. Thomas has to fall back on the general principle of analogy. What is interesting here is that when a metaphorical tradition conflicts with the metonymic need for conceptual and moral models, it is the tradition that has to give way.

Again, the AV represents Jesus as saying to Nicodemus in John 3:8,
"The wind bloweth where it listeth . . . so is everyone that is born of the Spirit." This is a metonymic translation: "Spirit" is a conception, identified with the Holy Spirit of Christian doctrine, and "wind" is a concrete illustration of it. But in the Greek text the same word, pneuma, is used for both wind and spirit. Hence a purely metaphorical translation is also possible: "The wind blows where it likes . . . that's what everyone is like who is born of the wind." We may find this rendering a trifle unsettling, and so, apparently, did Nicodemus, who heard only the word pneuma. But the example shows how deeply the history of language, and of thought in relation to language, is involved in translation.

We spoke of a
verbal magic in the metaphorical phase, arising from a sense of an energy common to words and things, though embodied and controlled in words. In the metonymic phase this sense of verbal magic is sublimated into a quasi-magic inherent in sequence or linear ordering. Hence the medieval fascination with the syllogism and the great medieval dream of deducing all knowledge from the premises of revelation. Later we have the "I think, therefore I am" of Descartes, where the operative word is "therefore," because before we can accept the proposition we must accept the cogency and reality of therefores. The Cartesian formula is close to being a restatement of the old ontological argument for God, which is reducible to "I think, there-fore God exists." Beliefs of this period that may seem to us perverse about, for example, predestination or the divine right of kings--may be stubbornly clung to because of the strength of the feeling: if you accept this, then you must, etc. During the Christian centuries, too, the fear of "heresy," or logical deviation from Christian premises, amounted to what was perhaps the deadliest social psychosis in history.

It was much earlier, however, that a
third phase of language had begun to develop out of a dissatisfaction with certain elements in second-phase language, two in particular. Syllogistic reasoning, it was felt, led to nothing genuinely new, because its conclusions were already contained within its premises, and so its march across reality seemed increasingly to be a verbal illusion. Then again, an analogical approach to language appeared to have no criteria for distinguishing existents from non-existents. Grammatically, logically, and syntactically, there is no difference between a lion and a unicorn: the question of actual existence does not enter the ordering of words as such. And if it does not, there can be no real difference between reasoning and rationalizing, as both procedures order words in the same way. The difference can be established only by criteria external to words, and the first of these criteria has to be that of "things," or objects in nature.

This third phase of language begins roughly in the sixteenth century, where it accompanies certain tendencies in the Renaissance and Reformation, and attains cultural ascendancy in the eighteenth.
In English literature it begins theoretically with Francis Bacon, and effectively with Locke. Here we start with a clear separation of subject and object, in which the subject exposes itself, in sense experience, to the impact of an objective world. The objective world is the order of nature; thinking or reflection follows the suggestions of sense experience, and words are the servomechanisms of reflection. Continuous prose is still employed, but all deductive procedures are increasingly subordinated to a primary inductive and fact-gathering process. The seventeenth-century poet Cowley, hailing Bacon as the Moses who had led modern thought out of the Egypt of superstition, says:

              From words, which are but pictures of the thought,
              (Though we our thoughts from them perversely drew)
              To things, the mind's right object, he it brought.

Hence this approach treats language as primarily descriptive of an objective natural order. The ideal to be achieved by words is framed on the model of truth by correspondence. A verbal structure is set up beside what it describes, and is called "true" if it seems to provide a satisfactory correspondence to it. The criterion of truth is related to the external source of the description rather than to the inner consistency of the argument. Its controlling figure, then, is a kind of simile: a true verbal structure is one that is like what it describes. In this phase we return to a direct relation between the order of nature and the order of words, as in the metaphorical phase, but with a sharp and consistent distinction between the two. This involves a reaction against the transcendental perspective of the second phase, and "impossibility forms of third-phase thinking demonstrate the "impossibility of metaphysics," or declare that all religious questions are unmeaning

The central principle of Locke, that nothing exists in the intellect that has not previously existed in the senses, had been an established axiom for manbefore him. But considerable social changes have to take place before this use of language becomes culturally dominant, and separable from other modes.

One of these changes is the growth of science on a basis of inductive observation. Science assumes two levels of sense perception: a particular accidental level that is largely illusion, and an ideal level that is our real source of knowledge.

The problem of illusion and reality therefore becomes a central in third-phase language. Copernicus is the great symbol for a new realization that such words as "sunrise" and "sunset," though metaphorically efficient, had become "only" metaphors, and that so far as they were descriptive, what they described was illusory. Darwin is the great symbol for a new realization that divine creation, as generally conceived, was an illusion projected from the evolutionary operations within nature. Einstein is the great symbol for a new realization that matter, which up to the twentieth century had been the great bastion of the objectivity of the world, was an illusion of energy. With this, however, the sense of the clear separation of subject and object, which was so marked a feature of the scientific attitude up to that point, overreached itself and began to come to an end. It was no longer possible to separate the observer from what he observes: the observer had to become an observed object too.

The thought suggests itself that
we may have completed a gigantic cycle of language from Homer's time, where the word evokes the thing, to our own day, where the thing evokes the word, and are now about to go around the cycle again, as we seem now to be confronted once again with an energy common to subject and object which can be expressed verbally only through some form of metaphor. It is true that many metaphorical elements are reappearing in our language, but it is rather the positive aspect of the same process--that we may be entering a new phase altogether in our understanding of language--that has to be kept in mind. Certainly it is interesting and rather reassuring that there should be so heavy an emphasis on language and linguistic models in contemporary thought, apart from whatever embodies the emphasis.

It is primarily to Roman Jakobson that we owe the distinction between the metaphorical and the metonymic, and I apologize for adding one more ingredient to what very quickly became a considerably overspiced stew.
It seems to me that there are three major senses in which the word "metonymic" can be used. First, it is a figure of speech in which an image is "put for" another image: this is really a species of metaphor. Second, it is a mode of analogical thinking and writing in which the verbal expression is "put for" something that by definition transcends adequate verbal expression: this is roughly the sense in which I use it. Third, it is a mode of thought and speech in which the word is "put for" the object it describes: this corresponds more or less to my "descriptive" phase. There are no rights and wrongs in such matters, but it seems to me useful to separate both the language of immanence, which is founded on metaphor, and the language of transcendence, which is founded on
metonymy in my sense, from descriptive language.

In the first, or metaphorical, phase of language, the unifying element of verbal expression is the "god," or personal nature-spirit. In the second phase the conception of a transcendent "God" moves into the center of the order of words. In the third phase the criterion of reality is the source of sense experience in the order of nature, where "God" is not to be found, and where "gods" are no longer believed in. Hence for the third phase of language the word "God" becomes linguistically unfunctional, except when confined to special areas outside its jurisdiction. Mythological space became separated from scientific space with the new astronomy of the seventeenth century, and mythological time from scientific time with nineteenth-century geology and biology. Both developments helped to push the conception of God out of the world of time and space, even as a hypothesis. The charge of "God-building" is a most damaging one to a third phase writer, and the subject that used to be called natural theology does not now make much cultural impact, with the remarkable exception of Teilhard de Chardin.

In the nineteenth century there were many thinkers, mainly of the idealistic school, who adhered to the metonymic tradition with its God. But some even of them give an impression of having said to themselves: Here's this word "God"; what am I to do with it? What they did was often ingenious, but frequently confirmed the feeling that the conception of God, like Biblical metaphors in metonymic theology, was becoming, however unconsciously, a cumbersome piece of traditional baggage. In a conception of language where no premises are beyond scrutiny, there is nothing to stop anyone from returning to square one and the question: Is there a God? What is significant about this is that the answer, if it is to remain within the framework of third-phase language, can only be no, because any question beginning with "is there" is, so to speak, already an ungodly question, and "a god" is for all practical purposes no God. Nietzsche's formula "God is dead," despite the amount of attention it has attracted, was incidental to his more important aim of de-deifying the natural environment, and in particular of removing the metaphor of "law" from ordinary consciousness to describe the operations of nature. There are no laws in nature, Nietzsche says, only necessities; but the metaphor "law of nature" carries with it a vestigial sense of a personality who commands and other personalities (ourselves) who have the option of obeying or disobeying; and this vestigial metaphor, for Nietzsche, is a superstition in the most exact sense of an inorganic survival of tradition.

The political and psychological aspects of third-phase writing led to similar positions.
One of the earliest of third-phase writers, Machiavelli, attempted to distinguish and isolate the tactical use of illusion in the art of ruling. For Rousseau civilization was largely an illusion concealing a society of nature and reason; for Marx the whole second-phase approach to language had become an ideology, or facade of ascendant-class authority; for Freud the language of consciousness was largely a screen concealing other motives for speech. To conservative thinkers, including Burke, the facade or -authority in society revealed the real structure of that society. There is no social contract. Burke maintained, except the contract that a society shows it to accepted by its structure.

What I am concerned with at present is
not the question whether God is dead or obsolete, but with the question of what resources of language may be dead or obsolete. The metaphorical and metonymic phases of language have been in large measure outgrown because of the obvious limitations that they imposed on the human mind. But it seems clear that the descriptive phase also has limitations, in a world where its distinction of subject and object so often does not work. There is no question of giving up descriptive language, only of relating it to a broader spectrum of verbal expression.

In Exodus 3: 14 , though God also gives himself a name, he defines himself (according to the AV) as "I am that I am," which scholars say is more accurately rendered "I will be what I will be." That is, we might come closer to what is meant in the Bible by the word "God" if we understood it as a verb, and not a verb of simple asserted existence but a verb implying a process accomplishing itself. This would involve trying to think our way back to a conception of language in which words were words of power, conveying primarily the sense of forces and energies rather than analogues of physical bodies. To some extent this would be a reversion to the metaphorical language of primitive communities, as our earlier references to a cycle of language and the "primitive" word mana suggested. But it would also be oddly contemporary with post-Einsteinian physics, where atoms and electrons are no longer thought of as things but rather as traces of processes. God may have lost his function as the subject of predicate, but may not be so much dead as entombed in a dead language.

The Biblical terms usually rendered "word," including the logos of the Gospel of John, are solidly rooted in the metaphorical phase of language, where the word was an element of creative power. According to Genesis 1:4, "God said, Let there be light; and there was light." That is, the word was the creative agent that brought the thing into being. This is usually thought of as characteristically Hebrew in approach, although in Heraclitus the term logos is also essentially metaphorical, and still expresses a unity of human consciousness and physical phenomena. In the metonymic phase logos takes on rather the meaning of an analogical use of words to convey the sense of a rational order. This order is thought of as antecedent to both consciousness and nature. Philo and the author of John combine the two traditions, and John's "In the beginning was the logos" is a New Testament commentary on the opening of Genesis, identifying the original creative word with Christ.

Erasmus, in the Latin translation appended to his edition of the Greek New Testament, renders In the beginning was the Word" as , "In principio erat sermo."
This is a purely metonymic translation: in the beginning, Erasmus assumes, was the infinite mind, with its interlocking thoughts and ideas out of which the creative words emerged. Erasmus is clearly more influenced than Jerome by the later Greek history of the word. It would be cheap parody to say that Erasmus really means "In the beginning was continuous prose," but the link between his "sermo" and the development of continious prose is there nonetheless. At the beginning of the third phase we have Goethe's Faust, who claims to have studied theology but seems not to understand it very well, struggling with the same phrase. He rejects "das Wort," and traverses the whole cycle of language as outlined above, passing through the second-phase "der Sinn," and emerging finally with "die That," the event or existential reality that words describe at secondhand. At that point Faust begins to fall into the power of Mephistopheles, the spirit of denial. What significance this has I am not sure, except that while it is not easy to translate "In the beginning was the Word," there seems to be no future in deliberately mistranslating it. Still, Faust makes us realize how completely we have lost the metaphorical clue to what John means by logos. For John goes on to say "And the logos became flesh." Evidently he thought of this as an intelligible statement of the, type "And the boy became a man," or "And the ice became water." But within a descriptive framework of language it can be only an unintelligible statement of the type "And the apple became an orange." For descriptive language, the word
has no power to be anything but aword.

Each of our three phases of language has a characteristic word for the human entity that uses the language. In the metaphorical phase, where the world is held together by a plurality of gods, there is often assumed to be a corresponding plurality of psychic forces that disin-tegrate or separate at death. Ancient Egypt had a ba and a ka and several other entities, besides the mummified body itself;
Homer (or a later editor) speaks of Hercules as existing after death simultaneously as a god in Olympus and a shade in Hades. Even Aristotle's De Anima describes a complex soul. But the nearest to the purely metaphorical conception is perhaps the word "spirit," which, with its overtones of "breath," expresses the unifying principle of life that gives man a participating energy with nature.

In proportion as
metonymic thinking and its monotheistic God developed, man came to be thought of as a single "soul" and a body, related by the metaphor of "in." Human consciousness feels that it is inside a body it knows next to nothing about, even such elementary facts as the circulation of the blood being relatively recent discoveries. Hence it cannot feel that the body is identical with consciousness: the body is born of nature and will return to nature, but the soul belongs to the transcendent world and will return to that world. The figures employed to describe their relation include a body in a tomb, a prisoner in a cell, a peasant in a decaying cottage, a bird in a cage, and the like. The separation of body and soul at death is thought of as a vertical one, the soul going "up" and the body "down."

In the third phase the conception associated with consciousness modulates from "soul" to "mind," and the relation with the bodily world of nature, including one's own body, becomes more horizontal. By this time the "mind" has become firmly located in the head, and
consciousness is in fact often thought of as a function of the brain.

All the languages relevant to the Bible distinguish between soul and spirit: Hebrew has nephesh and mach, Greek psyche and pneuma , Latin anima and spiritus; and there are similar distinctions in modern languages.
No one would claim that there was a consistent use of either word in the Bible, but neither would anyone speak of the third person of the Trinity as the Holy Soul, and Paul's prayer for his correspondent's "spirit and soul and body" (I Thessalonians 5:23) suggests that the difference between soul and spirit means something. Jesus' resurrection was a bodily one, and Paul explains (I Corinthians 15:44) that what enters the resurrection is not the soul or abstract essence of the body but a "spiritual body." This spiritual body is contrasted with the natural body, or "flesh and blood," but the phrase still suggests that immortality must include the body, in however transfigured a form, as it did in Jesus' resurrection. Here again the New Testament adheres to older metaphorical modes of thought rather than to the more up-to-date and rational Greek ones.

Christianity placed the doctrine of the resurrection of the body in its creed, though the addition seems historically to have had little effect on the soul-body dichotomy.
The people Dante meets in his visions of hell and purgatory and heaven are souls of the dead: at the Last Judgment, we are told, they go back to pick up their bodies, but what change that will make is very little emphasized, except that clearly it will make hell hurt a lot more. The term "spirit" seems to belong properly only to the Holy Spirit and, in a different context, to the angels: for man, and for discarnate beings like elemental "spirits," it seems to be a mere doublet of "soul." Yet Paul, again (I Corinthians 2:14), contrasts the pneumatikos or spiritual with the psychikos , the soul-body. The AV renders this latter term as the "natural" man: the difficulty in translation is that there is no English adjective related to "soul" corresponding to "spiritual." But Paul seems to be drawing the essential line between spirit and soul, not
between soul and body

Each phase of language has its characteristic virtues as well as its limitations.
In the first phase, language can be used with an immediacy and vitality, such as we find in Homer, that later ages never consistently recapture. Yet this use of language is restricted by an identity with nature from which metonymic dialectic has freed itself. The crossing of the bridge from "gods" to "God," which has already taken place in the Bible, is felt as a release from the tyranny of nature.