SELECTED

LITERARY

ESSAYS

by C.S.LEWIS

Edited by

WALTER HOOPER

CAMBRIDGE

At the University Press

1969

Published by the Syndics of the Cambridge University Press

Bentley House, 200 Euston Road, London n.w.i

American Branch: 32 East 57th Street, New York, N.Y.10022

This Selection, Preface and Notes © Cambridge University Press 1969

Library of Congress Catalogue Card Number: 74-85724

Standard Book Number: 521 0744t x

Printed in Great Britain

at the University Printing House, Cambridge

(Brooke Crutchley, University Printer)

Tie Librari

fe Beerst*?

Contents

preface by Walter Hooper

I De Descriptione Temporum

vi The Alliterative Metre

3 What Chaucer really did to // Filostrato

4 The Fifteenth-Century Heroic Line

5 Hero and Leander

6 Variation in Shakespeare and Others

u-7 Hamlet: The Prince or The Poem?

8 Donne and Love Poetry in the Seventeenth Century

9 The Literary Impact of the Authorised Version

10 The Vision of John Bunyan

II Addison

12 Four-Letter Words

13 A Note on Jane Austen

14 Shelley, Dryden, and Mr Eliot

15 Sir Walter Scott

16 William Morris

17 Kipling's World

18 Bluspels and Flalansferes: A Semantic Nightmare

19 High and Low Brows

20 Metre

21 Psycho-Analysis and Literary Criticism

22 The Anthropological Approach

INDEX

page vn

1

27

45

58

74

88

106

126

146

154

169

175

187

209

219

232

251

266

280

286

301

312

SELECTED LITERARY ESSAYS

his view of life is to go on from Kipling and to add the necessary-

correctives—not to deny what he has shown. After Kipling there is

no excuse for the assumption that all the important things in a man's

life happen between the end of one day's work and the beginning of

the next. There is no good putting on airs about Kipling. The things

he mistook for gods may have been only 'spirits of another sort'; but

they are real things and strong.

250

18

Bluspels and Flalansferes.

A Semantic Nightmare

We are often compelled to set up standards we cannot reach ourselves and to

lay down rules we could not ourselves satisfy.

lord Coleridge, c.j. (Law Reports, Queen's

Bench Division xiv, p. 288 in Reg. v. Dudley and Stephen)

Philologists often tell us that our language is full of dead metaphors.

In this sentence, the word 'dead' and the word !metaphors' may turn

out to be ambiguous; but the fact, or group of facts, referred to, is

one about which there is no great disagreement. We all know in a

rough and ready way, and all admit, these things which are being

called 'dead metaphors', and for the moment I do not propose to

debate the propriety of the name. But while their existence is not

disputed, their nature, and their relation to thought, gives rise to a

great deal of controversy. For the benefit of any who happen to have

avoided this controversy hitherto, I had better make plain what it is,

by a concrete example. Bréal in his Semantics1 often spoke in meta-

phorical, that is consciously, rhetorically, metaphorical language, of

language itself. Messrs Ogden and Richards in The Meaning of

Meaning took Bréal to task on the ground that ' it is impossible thus

to handle a scientific matter in metaphorical terms'.2 Barfield in his

Poetic Diction retorted that Ogden and Richards were, as a matter of

fact, just as metaphorical as Bréal. They had forgotten, he complained,

that all language has a figurative origin and that the 'scientific' terms

on which they piqued themselves—words like organism, stimulus,

reference—were not miraculously exempt. On the contrary, he main-

tained, 'those who profess to eschew figurative expressions are really

confining themselves to one very old kind of figure'—'they are

absolutely rigid under the spell of those verbal ghosts of the physical

sciences, which today make up practically the whole meaning-system

1 M. J. A. Bréal, Semantics: studies in the science of meaning, trans. Mrs Henry Cust, with

a Preface by J. P. Postgate (London, 1900).

2 C. K. Ogden and I. A. Richards, The Meaning of Meaning (London, 1923), pp. 4-5.

SELECTED LITERARY ESSAYS

of so many European minds'.* Whether Ogden and Richards will see

fit, or have seen fit, to reply to this, I do not know; but the lines on

which any reply would run are already traditional. In fact the whole

debate may be represented by a very simple dialogue.

A. You are being metaphorical.

B. You are just as metaphorical as I am, but you don't know it.

A. No, I'm not. Of course I know all about attending once having

meant stretching, and the rest of it. But that is not what it

means now. It may have been a metaphor to Adam—but I

am not using it metaphorically. What I mean is a pure concept

with no metaphor about it at all. The fact that it was a metaphor

is no more relevant than the fact that my pen is made of wood.

You are simply confusing derivation with meaning.

There is clearly a great deal to be said for both sides. On the one

hand it seems odd to suppose that what we mean is conditioned by a

dead metaphor of which we may be quite ignorant. On the other

hand, we see from day to day, that when a man uses a current and

admitted metaphor without knowing it, he usually gets led into

nonsense; and when, we are tempted to ask, does a metaphor become

so old that we can ignore it with impunity? It seems harsh to rule that

a man must know the whole semantic history of every word he uses

—a history usually undiscoverable—or else talk without thinking.

And yet, on the other hand, an obstinate suspicion creeps in that we

cannot entirely jump off our own shadows, and that we deceive

ourselves if we suppose that a new and purely conceptual notion of

attention has replaced and superseded the old metaphor of stretching.

Here, then, is the problem which I want to consider. How far, if at

all, is thinking limited by these dead metaphors? Is Anatole France in

any sense right when he reduces 'The soul possesses God' to 'the

breath sits on the bright sky'? Or is the other party right when it

urges 'Derivations are one thing. Meanings are another'? Or is the

truth somewhere between them?

The first and easiest case to study is that in which we ourselves

invent a new metaphor. This may happen in one of two ways. It

may be that when we are trying to express clearly to ourselves or to

others a conception which we have never perfectly understood, a

new metaphor simply starts forth, under the pressure of composition

or argument. When this happens, the result is often as surprising and

* Owen Barfield, Poetic Diction: A Study in Meaning (London, 1928), p. 140.

252

BLUSPELS AND FLALANSFERES

illuminating to us as to our audience; and I am inclined to think that

this is what happens with the great, new metaphors of the poets. And

when it does happen, it is plain that our new understanding is bound

up with the new metaphor. In fact, the situation is for our purpose

indistinguishable from that which arises when we hear a new metaphor

from others; and for that reason, it need not be separately discussed.

One of the ways, then, in which we invent a new metaphor, is by

finding it, as unexpectedly as we might find it in the pages of a book;

and whatever is true of the new metaphors that we find in books will

also be true of those which we reach by a kind of lucky chance, or

inspiration. But, of course, there is another way in which we invent

new metaphors. When we are trying to explain, to some one younger

or less instructed than ourselves, a matter which is already perfectly

clear in our own minds, we may deliberately, and even painfully,

pitch about for the metaphor that is likely to help him. Now when

this happens, it is quite plain that our thought, our power of meaning,

is not much helped or hindered by the metaphor that we use. On the

contrary, we are often acutely aware of the discrepancy between our

meaning and our image. We know that our metaphor is in some

respects misleading; and probably, if we have acquired the tutorial

shuffle, we warn our audience that it is 'not to be pressed'. It is

apparently possible, in this case at least, to use metaphor and yet to

keep our thinking independent of it. But we must observe that it is

possible, only because we have other methods of expressing the same

idea. We have already our own way of expressing the thing: we could

say it, or we suppose that we could say it, literally instead. This clear

conception we owe to other sources—to our previous studies. We can

adopt the new metaphor as a temporary tool which we dominate and

by which we are not dominated ourselves, only because we have

other tools in our box.

Let us now take the opposite situation—that in which it is we

ourselves who are being instructed. I am no mathematician; and some

one is trying to explain to me the theory that space is finite. Stated

thus, the new doctrine is, to me, meaningless. But suppose he proceeds

as follows.

'You', he may say, 'can intuit only three dimensions; you therefore

cannot conceive how space should be limited. But I think I can show

you how that which must appear infinite in three dimensions, might

nevertheless be finite in four. Look at it this way. Imagine a race of

253

SELECTED LITERARY ESSAYS

people who knew only two dimensions—like the Flatlanders.1 And

suppose they were living on a globe. They would have no conception,

of course, that the globe was curved—for it is curved round in that

third dimension of which they have no inkling. They will therefore

imagine that they are living on a plane; but they will soon find out

that it is a plane which nowhere comes to an end; there are no edges

to it. Nor would they be able even to imagine an edge. For an edge

would mean that, after a certain point, there would be nothing to

walk on; nothing below their feet. But that below and above dimension

is just what their minds have not got; they have only backwards and

forwards, and left and right. They would thus be forced to assert that

their globe, which they could not see as a globe, was infinite. You can

see perfectly well that it is finite. And now, can you not conceive that

as these Flatlanders are to you, so you might be to a creature that

intuited four dimensions? Can you not conceive how that which seems

necessarily infinite to your three-dimensional consciousness might

none the less be really finite? ' The result of such a metaphor on my

mind would be—in fact, has been—that something which before was

sheerly meaningless acquires at least a faint hint of meaning. And if

the particular example does not appeal to every one, yet every one has

had experiences of the same sort. For all of us there are things which

we cannot fully understand at all, but of which we can get a faint

inkling by means of metaphor. And in such cases the relation between

the thought and the metaphor is precisely the opposite of the relation

which arises when it is we ourselves who understand and then invent

the metaphors to help others. We are here entirely at the mercy of

the metaphor. If our instructor has chosen it badly, we shall be

thinking nonsense. If we have not got the imagery clearly before us,

we shall be thinking nonsense. If we have it before us without knowing

that it is metaphor—if we forget that our Flatlanders on their globe

are a copy of the thing and mistake them for the thing itself—then

again we shall be thinking nonsense. What truth we can attain in

such a situation depends rigidly on three conditions. First, that the

imagery should be originally well chosen; secondly, that we should

apprehend the exact imagery; and thirdly that we should know that

the metaphor is a metaphor. (That metaphors misread as statements

of fact are the source of monstrous errors need hardly be pointed out.)

1 The inhabitants in the book by 'A Square' [Edwin A. Abbott], Flatland. A romance of

many dimensions (London, 1884).

254

BLUSPELS AND FLALANSFERES

I have now attempted to show two different kinds of metaphorical

situation as they are at their birth. They are the two extremes, and

furnish the limits within which our inquiry must work. On the one

hand, there is the metaphor which we invent to teach by; on the other,

the metaphor from which we learn. They might be called the Master's

metaphor, and the Pupil's metaphor. The first is freely chosen; it is

one among many possible modes of expression; it does not at all

hinder, and only very slightly helps, the thought of its maker. The

second is not chosen at all; it is the unique expression of a meaning

that we cannot have on any other terms; it dominates completely the

thought of the recipient; his truth cannot rise above the truth of the

original metaphor. And between the Master's metaphor and the Pupil's

there comes, of course, an endless number of types, dotted about in

every kind of intermediate position. Indeed, these Pupil-Teachers'

metaphors are the ordinary stuff of our conversation. To divide them

into a series of classes and sub-classes and to attempt to discuss these

separately would be very laborious, and, I trust, unnecessary. If we

can find a true doctrine about the two extremes, we shall not be at a

loss to give an account of what falls between them. To find the truth

about any given metaphorical situation will merely be to plot its

position. In so far as it inclines to the 'magistral' extreme, so far our

thought will be independent of it; in so far as it has a 'pupillary'

element, so far it will be the unique expression, and therefore the iron

limit of our thinking. To fill in this framework would be, as Aristotle

used to say, 'anybody's business'.

Our problem, it will be remembered, was the problem of'dead' or

'forgotten' metaphors. We have now gained some light on the

relation between thought and metaphor as it is at the outset, when the

metaphor is first made; and we have seen that this relation varies

greatly according to what I have called the 'metaphorical situation'.

There is, in fact, one relation in the case of the Master's metaphor,

and an almost opposite relation in that of the Pupil's metaphor.

The next step must clearly be to see what becomes of these two

relations as the metaphors in question progress to the state of death or

fossilization.

The question of the Master's Metaphor need not detain us long. I

may attempt to explain the Kantian philosophy to a pupil by the

following metaphor. 'Kant answered the question "How do I know

that whatever comes round the corner will be blue?" by the sup-

255

SELECTED LITERARY ESSAYS

position "lam wearing blue spectacles." ' In time I may come to use

'the blue spectacles' as a kind of shorthand for the whole Kantian

machinery of the categories and forms of perception. And let us

suppose, for the sake of analogy with the real history of language,

that I continue to use this expression long after I have forgotten the

metaphor which originally gave rise to it. And perhaps by this time the

form of the word will have changed. Instead of the 'blue spectacles'

I may now talk of the bloospel or even the bluspel. If I live long enough

to reach my dotage I may even enter on a philological period in

which I attempt to find the derivation of this mysterious word. I

may suppose that the second element is derived from the word spell

and look back with interest on the supposed period when Kant

appeared to me to be magical; or else, arguing that the whole word is

clearly formed on the analogy oí gospel, may indulge in unhistorical

reminiscences of the days when the Critique1 seemed to me irrefragably

true. But how far, if at all, will my thinking about Kant be affected by

all this linguistic process? In practice, no doubt, there will be some

subtle influence; the mere continued use of the word bluspel may have

led me to attribute to it a unity and substantiality which I should have

hesitated to attribute to 'the whole Kantian machinery of the categories

and forms of perception'. But that is a result rather of the noun-making

than of the death of the metaphor. It is an interesting fact, but hardly

relevant to our present inquiry. For the rest, the mere forgetting of

the metaphor does not seem to alter my thinking about Kant, just as

the original metaphor did not limit my thinking about Kant; provided

always—and this is of the last importance—that it was, to begin with,

a genuine Master's metaphor. I had my conception of Kant's philo-

sophy before I ever thought of the blue spectacles. If I have continued

philosophical studies I have it still. The 'blue spectacles' phrase was

from the first a temporary dress assumed by my thought for a special

purpose, and ready to be laid aside at my pleasure; it did not penetrate

the thinking itself, and its subsequent history is irrelevant. To any

one who attempts to refute my later views on Kant by telling me that

I don't know the real meaning of bluspel, I may confidently retort

' Derivations aren't meanings.' To be sure, if there was any pupillary

element in its original use, if I received, as well as gave, new under-

standing when I used it, then the whole situation will be different.

1 Immanuel Kant, Critique of Practical Reason, and other works on the Theory of Ethics,

trans. T. K. Abbott (London, 1879).

256

BLUSPELS AND FLALANSFERES

And it is fair to admit that in practice very few metaphors can be

purely magistral; only that which to some degree enlightens ourselves

is likely to enlighten others. It is hardly possible that when I first used

the metaphor of the blue spectacles I did not gain some new awareness

of the Kantian philosophy; and, so far, it was not purely magistral.

But I am deliberately idealizing for the sake of clarity. Purely magistral

metaphor may never occur. What is important for us is to grasp that

just in so far as any metaphor began by being magistral, so far I can

continue to use it long after I have forgotten its metaphorical nature,

and my thinking will be neither helped nor hindered by the fact that

it was originally a metaphor, nor yet by my forgetfulness of that fact.

It is a mere accident. Here, derivations are irrelevant to meanings.

Let us now turn to the opposite situation, that of the Pupil's

Metaphor. And let us continue to use our old example of the unmathe-

matical man who has had the finitude of space suggested to him (we

can hardly say 'explained') by the metaphor of the Flatlanders on

their sphere. The question here is rather more complicated. In the

case of the Master's metaphor, by hypothesis, the master knew, and

would continue to know, what he meant, independently of the

metaphor. In the present instance, however, the fossilization of the

metaphor may take place in two different ways. The pupil may himself

become a mathematician, or he may remain as ignorant of mathematics

as he was before; and in either case, he may continue to use the

metaphor of the Flatlanders while forgetting its real content and its

metaphorical nature.

I will take the second possibility first. From the imagery of the

Flatlanders' sphere I have got my first inkling of the new meaning.

My thought is entirely conditioned by this imagery. I do not apprehend

the thing at all, except by seeing 'it could be something like this'.

Let us suppose that in my anxiety to docket this new experience, I

label the inkling or vague notion 'the Flatlanders' sphere'. When I

next hear the fourth dimension spoken of, I shall say, 'Ah yes—the

Flatlanders' sphere and all that.' In a few years (to continue our

artificial parallel) I may be talking glibly of the Flalansfere and may

even have forgotten the whole of the imagery which this word once

represented. And I am still, according to the hypothesis, profoundly

ignorant of mathematics. My situation will then surely be most

ridiculous. The meaning of Flalansfere I never knew except through

the imagery. I could get beyond the imagery, to that whereof the

257

SELECTED LITERARY ESSAYS

imagery was a copy, only by learning mathematics; but this I have

neglected to do. Yet I have lost the imagery. Nothing remains, then,

but the conclusion that the word Flalansfere is now really meaning-

less. My thinking, which could never get beyond the imagery, at

once its boundary and its support, has now lost that support. I mean

strictly nothing when I speak of the Flalansfere. I am only talking,

not thinking, when I use the word. But this fact will be long concealed

from me because Flalansfere, being a noun, can be endlessly fitted

into various contexts so as to conform to syntactical usage and to

give an appearance of meaning. It will even conform to the logical

rules; and I can make many judgements about the Flalansfere; such

as it is what it is, and has attributes (for otherwise of course it wouldn't

be a thing, and if it wasn't a thing, how could I be talking about it?),

and is a substance (for it can be the subject of a sentence). And what

affective overtones the word may have taken on by that time it is

dangerous to predict. It had an air of mystery from the first: before

the end I shall probably be building temples to it, and exhorting my

countrymen to fight and die for the Flalansfere. But the Flalansfere,

when once we have forgotten the metaphor, is only a noise.

But how if I proceed, after once having grasped the metaphor of

the Flatlanders, to become a mathematician? In this case, too, I may

well continue to use the metaphor, and may corrupt it in form till it

becomes a single noun, the Flalansfere. But I shall have advanced, by

other means, from the original symbolism; and I shall be able to study

the thing symbolized without reference to the metaphor that first

introduced me to it. It will then be no harm though I should forget

that Flalansfere had ever been metaphorical. As the metaphor, even if

it survived, would no longer limit my thoughts, so its fossilization

cannot confuse them.

The results which emerge may now be summarized as follows. Our

thought is independent of the metaphors we employ in so far as these

metaphors are optional: that is, in so far as we are able to have the

same idea without them. For that is the real characteristic both of

the magistral metaphors and of those which become optional, as the

Flatlanders would become, if the pupil learned mathematics. On the

other hand, where the metaphor is our only method of reaching a

given idea at all, there our thinking is limited by the metaphor so

long as we retain the metaphor; and when the metaphor becomes

fossilized, our 'thinking' is not thinking at all, but mere sound or

258

BLUSPELS AND FLALANSFERES

mere incipient movements in the larynx. "We are now in a position

to reply to the statement that 'Derivations are not meanings', and to

claim that 'we know what we mean by words without knowing the

fossilized metaphors they contain'. We can see that such a statement,

as it stands, is neither wholly true nor wholly false. The truth will

vary from word to word, and from speaker to speaker. No rule of

thumb is possible, we must take every case on its merits. A word

can bear a meaning in the mouth of a speaker who has forgotten its

hidden metaphor, and a meaning independent of that metaphor, but

only on certain conditions. Either the metaphor must have been

optional from the beginning, and have remained optional through all

the generations of its use, so that the conception has always used and

still uses the imagery as a mere tool; or else, at some period subsequent

to its creation, we must have gone on to acquire, independently of the

metaphor, such new knowledge of the object indicated by it as enables

us now, at least, to dispense with it. To put the same thing in another

way, meaning is independent of derivation only if the metaphor was

originally 'magistral'; or if, in the case of an originally pupillary

metaphor, some quite new kind of apprehension has arisen to replace

the metaphorical apprehension which has been lost. The two conditions

may be best illustrated by a concrete example. Let us take the word for

soul as it exists in the Romance language. How far is a man entitled

to say that what he means by the word âme or anima is quite inde-

pendent of the image of breathing, and that he means just the same

(and just as much) whether he happens to know that 'derivation' or

not? We can only answer that it depends on a variety of things. I

will enumerate all the formal possibilities for the sake of clearness : one

of them, of course, is too grotesque to appear for any other purpose.

1. The metaphor may originally have been magistral. Primitive

men, we are to suppose, were clearly aware, on the one hand, of an

entity called soul; and, on the other, of a process or object called

breath. And they used the second figuratively to suggest the first—

presumably when revealing their wisdom to primitive women and

primitive children. And we may suppose, further, that this magistral

relation to the metaphor has never been lost: that all generations, from

the probably arboreal to the man saying 'Blast your soul' in a pub

this evening, have kept clearly before them these two separate entities,

and used the one metaphorically to denote the other, while at the

same time being well able to conceive the soul unmetaphorically, and

259

SELECTED LITERARY ESSAYS

using the metaphor merely as a colour or trope which adorned but

did not influence their thought. Now if all this were true, it would

unquestionably follow that when a man says anima his meaning is

not affected by the old image of breath; and also, it does not matter

in the least whether he knows that the word once suggested that

image or not. But of course all this is not true.

2. The metaphor may originally have been pupillary. So far from

being a voluntary ornament or pedagogic device, the ideas of breath

or something like breath may have been the only possible inkling that

our parents could gain of the soul. But if this was so, how does the

modern user of the word stand? Clearly, if he has ceased to be aware

of the metaphorical element in anima, without replacing the meta-

phorical apprehension by some new knowledge of the soul, borrowed

from other sources, then he will mean nothing by it; we must not, on

that account, suppose that he will cease to use it, or even to use it (as

we say) intelligibly—i.e. to use it in sentences constructed according

to the laws of grammar, and to insert these sentences into those

conversational and literary contexts where usage demands their

insertion. If, on the other hand, he has some independent knowledge

of the entity which our ancestors indicated by their metaphor of

breath, then indeed he may mean something.

I take it that it is this last situation in which we commonly suppose

ourselves to be. It doesn't matter, we would claim, what the majestic

root GNA really stood for: we have learned a great deal about knowing

since those days, and it is these more recent acquisitions that we use

in our thinking. The first name for a thing may easily be determined

by some inconsiderable accident. As we learn more, we mean more;

the radical meaning of the old syllables does not bind us; what we

have learned since has set us free. Assuredly, the accident which led

the Romans to call all Hellenes Graeci did not continue to limit their

power of apprehending Greece. And as long as we are dealing with

sensible objects this view is hardly to be disputed. The difficulty begins

with objects of thought. It may be stated as follows.

Our claim to independence of the metaphor is, as we have seen, a

claim to know the object otherwise than through that metaphor. If

we can throw the Flatlanders overboard and still think the fourth

dimension, then, and not otherwise, we can forget what Flalansfere

once meant and still think coherently. That was what happened, you

will remember, to the man who went on and learned mathematics. He

260

BLUSPELS AND FLALANSFERES

came to apprehend that of which the Flatlanders' sphere was only the

image, and consequently was free to think beyond the metaphor and

to forget the metaphor altogether. In our previous account of him,

however, we carefully omitted to draw attention to one very remark-

able fact: namely, that when he deserted metaphor for mathematics

he did not really pass from symbol to symbolized, but only from one

set of symbols to another. The equations and what-nots are as unreal

as metaphorical, if you like, as the Flatlanders' sphere. The mathe-

matical problem I need not pursue further; we see at once that it

casts a disquieting light on our linguistic problem. We have hitherto

been speaking as if we had two methods of thought open to us- the

metaphorical, and the literal. We talked as if the creator of a magistral

metaphor had it always in his power to think the same concept literally

if he chose. We talked as if the present-day user of the word anima

could prove his right to neglect that word's buried metaphor by

turning round and giving us an account of the soul which was not

metaphorical at all. That he has power to dispense with the particular

metaphor of breath, is of course agreed. But we have not yet inquired

what he can substitute for it. If we turn to those who are most anxious

to tell us about the soul—I mean the psychologists—we shall find that

the word anima has simply been replaced by complexes, repressions

censors, engrams, and the like. In other words the breath has been

exchanged for tyings-up, shovings-back, Roman magistrates, and

scratchmgs. If we inquire what has replaced the metaphorical bright

sky of primitive theology, we shall only get a perfect substance, that

is, a completely made lying-under, or—which is very much better, but

equally metaphorical—a universal Father, or perhaps (in English) a

loaf-carver, in Latin a householder, in Romance a person older than.

The point need not be laboured. It is abundantly clear that the freedom

from a given metaphor which we admittedly enjoy in some cases is

often only a freedom to choose between that metaphor and others.

Certain reassurances may, indeed, be held out. -In the first place

our distinction between the different kinds of metaphorical situation

can stand; though it is hardly so important as we had hoped. To have

a choice of metaphors (as we have in some cases) is to know more

than we know when we are the slaves of a unique metaphor. And, in

the second place, all description or identification, all direction of our

own thought or another's, is not so metaphorical as definition. If,

when challenged on the word anima, we proceed to define, we shall

261

SELECTED LITERARY ESSAYS

only reshuffle the buried metaphors; but if we simply say (or think)

'what I am', or 'what is going on in here', we shall have at least

something before us which we do not know by metaphor. We shall

at least be no worse off than the arboreal psychologists. At the same

time, this method will not really carry us far. "What's going on here'

is really the content of haec anima: for anima we want ' The sort of

thing that is going on here', and once we are committed to sorts and

kinds we are adrift among metaphors.

We have already said that when a man claims to think independently

of the buried metaphor in one of his words, his claim may sometimes

be allowed. But it was allowed only in so far as he could really supply

the place of that buried metaphor with new and independent appre-

hension of his own. We now see that this new apprehension will

usually turn out to be itself metaphorical; or else, what is very much

worse, instead of new apprehension we shall have simply words—each

word enshrining one more ignored metaphor. For if he does not know

the history of anima, how should he know the history of the equally

metaphorical words in which he defines it, if challenged? And if he

does not know their history and therefore their metaphors, and if he

cannot define them without yet further metaphors, what can his

discourse be but an endless ringing of the changes on such bluspels

and Flalansferes as seem to mean, indeed, but do not mean? In reality,

the man has played us a very elementary trick. He claimed that he

could think without metaphor, and in ignorance of the metaphors

fossilized in his words. He made good the claim by pointing to the

knowledge of his object which he possessed independently of the

metaphor; and the proof of this knowledge was the definition or

description which he could produce. We did not at first observe that

where we were promised a freedom from metaphor we were given only

a power of changing the metaphors in rapid succession. The things he

speaks of he has never apprehended literally. Yet only such genuinely

literal apprehension could enable him to forget the metaphors which

he was actually using and yet to have a meaning. Either literalness, or

else metaphor understood: one or other of these we must have; the

third alternative is nonsense. But literalness we cannot have. The man

who does not consciously use metaphors talks without meaning. We

might even formulate a rule: the meaning in any given composition is

in inverse ratio to the author's belief in his own literalness.

If a man has seen ships and the sea, he may abandon the metaphor

262

BLUSPELS AND FLALANSFERES

of a sea-stallion and call a boat a boat. But suppose a man who has

never seen the sea, or ships, yet who knows of them just as much as

he can glean, say from the following list of Kenningar—sea-stallions,

winged-logs, wave-riders, ocean-trains. If he keeps all these together

in his mind, and knows them for the metaphors they are, he will be

able to think of ships, very imperfectly indeed, and under strict

limits, but not wholly in vain. But if instead of this he pins his faith

on the particular kenning, ocean-trains, because that kenning, with its

comfortable air of machinery, seems to him somehow more safely

prosaic, less flighty and dangerous than its fellows, and if, contracting

that to the form oshtrans, he proceeds to forget that it was a metaphor,

then, while he talks grammatically, he has ceased to think of anything.

It will not avail him to stamp his feet and swear that he is literal; to

say 'An oshtran is an oshtran, and there's an end. I mean what I mean.

What I mean is what I say.'

The remedy lies, indeed, in the opposite direction. When we pass

beyond pointing to individual sensible objects, when we begin to

think of causes, relations, of mental states or acts, we become incurably

metaphorical. We apprehend none of these things except through

metaphor: we know of the ships only what the Kenningar will tell us.

Our only choice is to use the metaphors and thus to think something,

though less than we could wish; or else to be driven by unrecognized

metaphors and so think nothing at all. I myself would prefer to em-

brace the former choice, as far as my ignorance and laziness allow me.

To speak more plainly, he who would increase the meaning and

decrease the meaningless verbiage in his own speech and writing, must

do two things. He must become conscious of the fossilized metaphors

in his words; and he must freely use new metaphors, which he creates

for himself. The first depends upon knowledge, and therefore on

leisure; the second on a certain degree of imaginative ability. The

second is perhaps the more important of the two: we are never less

the slaves of metaphor than when we are making metaphor, or hearing

it new made. When we are thinking hard of the Flatlanders, and at

the same time fully aware that they are a metaphor, we are in a

situation almost infinitely superior to that of the man who talks of the

Flalansfere and thinks that he is being literal and straightforward.

If our argument has been sound, it leads us to certain rather re-

markable conclusions. In the first place it would seem that we must

be content with a very modest quantity of thinking as the core of all

263

SELECTED LITERARY ESSAYS

our talking. I do not wish to exaggerate our poverty. Not all our

words are equally metaphorical, not all our metaphors are equally

forgotten. And even where the old metaphor is lost there is often a

hope that we may still restore meaning by pointing to some sensible

object, some sensation, or some concrete memory. But no man can

or will confine his cognitive efforts to this narrow field. At the very

humblest we must speak of things in the plural; we must point not

only to isolated sensations, but to groups and classes of sensations;

and the universal latent in every group and every plural inflection

cannot be thought without metaphor. Thus far beyond the security

of literal meaning all of us, we may be sure, are going to be driven

by our daily needs; indeed, not to go thus far would be to abandon

reason itself. In practice we all really intend to go much farther. Why

should we not? We have in our hands the key of metaphor, and it

would be pusillanimous to abandon its significant use, because we

have come to realize that its meaningless use is necessarily prevalent.

We must indeed learn to use it more cautiously; and one of the chief

benefits to be derived from our inquiry is the new standard of

criticism which we must henceforward apply both to our own

apparent thought and to that of others. We shall find, too, that real

meaning, judged by this standard, does not come always where we

have learned to expect. Flalansfere and Muspels will clearly be most

prevalent in certain types of writers. The percentage of mere syntax

masquerading as meaning may vary from something like ioo per cent,

in political writers, journalists, psychologists, and economists, to

something like forty per cent, in the writers of children's stories.

Some scientists will fare better than others: the historian, the

geographer, and sometimes the biologist will speak significantly more

often than their colleagues; the mathematician, who seldom forgets

that his symbols are symbolic, may often rise for short stretches to

ninety per cent, of meaning and ten of verbiage. The philosophers

will differ as widely from one another as any of the other groups

differ among themselves: for a good metaphysical library contains at

once some of the most verbal, and some of the most significant

literature in the world. Those who have prided themselves on being

literal, and who have endeavoured to speak plainly, with no mystical

tomfoolery, about the highest abstractions, will be found to be among

the least significant of writers: I doubt if we shall find more than a

beggarly five per cent, of meaning in the pages of some celebrated

264

BLUSPELS AND FLALANSFERES

'tough-minded' thinkers, and how the account of Kant or Spinoza

stands, none knows but heaven. But open your Plato, and you will

find yourself among the great creators of metaphor, and therefore

among the masters of meaning. If we turn to Theology—or rather to

the literature of religion—the result will be more surprising still; for

unless our whole argument is wrong, we shall have to admit that a

man who says heaven and thinks of the visible sky is pretty sure to

mean more than a man who tells us that heaven is a state of mind. It

may indeed be otherwise; the second man may be a mystic who is

remembering and pointing to an actual and concrete experience of

his own. But it is long, long odds. Bunyan and Dante stand where

they did; the scale of Bishop Butler, and of better men than he, flies

up and kicks the beam.

It will have escaped no one that in such a scale of writers the poets

will take the highest place; and among the poets those who have at

once the tenderest care for old words and the surest instinct for the

creation of new metaphors. But it must not be supposed that I am in

any sense putting forward the imagination as the organ of truth. We

are not talking of truth, but of meaning: meaning which is the ante-

cedent condition both of truth and falsehood, whose antithesis is not

error but nonsense. I am a rationalist. For me, reason is the natural

organ of truth; but imagination is the organ of meaning. Imagination,

producing new metaphors or revivifying old, is not the cause of

truth, but its condition. It is, I confess, undeniable that such a view

indirectly implies a kind of truth or tightness in the imagination

itself. I said at the outset that the truth we won by metaphor could

not be greater than the truth of the metaphor itself; and we have seen

since that all our truth, or all but a few fragments, is won by meta-

phor. And thence, I confess, it does follow that if our thinking is

ever true, then the metaphors by which we think must have been

good metaphors. It does follow that if those original equations,

between good and light, or evil and dark, between breath and soul

and all the others, were from the beginning arbitrary and fanciful—if

there is not, in fact, a kind of psycho-physical parallelism (or more)

in the universe—then all our thinking is nonsensical. But we cannot,

without contradiction, believe it to be nonsensical. And so, admittedly,

the view I have taken has metaphysical implications. But so has every

view.

265

I

High and Low Brows

Quick, quick.—Fling Peregrine Pickle under the toilet—throw Roderick Random

into the closet—put The Innocent Adultery into The Whole Duty of Man—cram

Ovid behind the bolster—there—put The Man of Feeling into your pocket—so,

so—now lay Mrs Chapone in sight, and leave Fordyce's Sermons open on the

table. SHERIDAN

Aristotle often begins his argument with what he calls an Isagoge, a col-

lection of instances which is not, if I understand the matter, intended (like

Mill's induction) to prove a general principle, but merely to open our

eyes to it. The following instances are meant to form such an isagoge.

i. Not many years ago a lady whose studies I was attempting to

supervise, propounded a literary theory of general application which

I found myself unable to accept. Applying the elenchus after my

fashion I inquired whether her theory would cover The Tale of Peter

Rabbit. After a silence of some minutes, she asked me if I thought

there was any use in introducing such an example into a serious

literary discussion. I replied that Peter Rabbit was a book and certainly

not so bad a book that it could be left outside the classification

'literature'. The lady, who is as honest as she is learned (and whom I

mention here with all respect), was not prepared to call Peter Rabbit

'bad'. 'Trivial' was the word she finally fixed on. But she was quite

sure that doctrines about 'literature' need not apply to it.

2.1 have heard of a preparatory school where the library regulations

divide the contents of the library into two classes: Good Books and

Books. The boys are allowed to take out two Good Books for every

one Book. To read a Good Book is meritorious, to read a Book only

tolerable. At the same time, those responsible for the regulations have

hesitated to label as 'bad' the books which they thus contrast with the

'good'.

3. I have heard the Head of a great college* praise the novels of

Anthony Hope and conclude by declaring with enthusiasm, 'They

* I have learned since that I misunderstood him, but as my Isagoge is meant for pure

illustration, not proof, I have thought it no dishonesty to let the example stand.

266

HIGH AND LOW BROWS

are the best "bad" books I've ever read.' Here, it will be seen, the

word 'bad' is actually used but used in a sense which admits, inside

the class Bad, distinctions of good, better, and best.

4. I have often heard—and who has not?—a plain man praise even

to rapture the delightful merits of some favourite story and end with

the humble reservation, ' Of course, I know it's not real literature.'

I trust that these four instances are already making clear to you

what it is I want to discuss. In all of them we see a distinction made

between two kinds of book, to the one of which a certain honour is

attached, and to the other a certain note of ignominy. Yet in spite of

this there is a reluctance to identify this distinction with the plain

distinctions of good from bad or better from worse. Those who uphold

the distinction prefer to call the inferior class popular, common,

commercial, cheap, trashy, or the like, and the superior class literary,

classical, serious, or artistic. In our own time the two odious adjectives

Lowbrow and Highbrow have been introduced as the names of the

two classes and bid fair to oust all their rivals. There will also be

noticed in the first, third, and fourth of my instances a suggestion

that the lowbrow works are so different in kind from the highbrow

that they have a goodness and badness of their own, a servile virtue

and servile vice peculiar to themselves, that they are to be judged by

peculiar standards, and that what is said of literature simpliciter is not

said of them.

Now it seems to me that in this popular distinction there is some

confusion between degrees and kinds of merit. If the lowbrow books

are really a special kind of book I do not see how they can be inferior

to the high. You cannot be beaten by a man unless you enter the same

competition with him, nor overtaken by a man (as Chesterton

observes) unless you both run in the same direction. At present the

distinction is certainly used to allow us the satisfaction of despising

certain authors and readers without imposing on us the labour of

showing that they are bad. It is also used, I find, to allow people to

enjoy lowbrow art without gratitude on the one hand or shame on

the other, and those who would hesitate to say, 'Let's go and see

something bad' will cheerfully say 'Let's go to a lowbrow film.' The

whole distinction seems to be made in order to enable us to have it

both ways.

In the following paper I propose two questions. (1) Is the class of

lowbrow books (or 'books' simply as at the Preparatory School)

10 267 HLL

SELECTED LITERARY ESSAYS

really the class of bad books? (2) If not, is the distinction useful in

some other way?1

As soon as we approach the first question we notice that even if

all the lowbrow books—which I am going to call Class A—are in

fact bad, even so the distinction between lowbrow and highbrow or

between class A and class B—will not coincide with the distinction

between bad books and good books, for the very obvious reason

that class B contains bad books too. Gorbuduc and Glover's Leónidas

and Dyer's Fleece, Gabriel Harvey's hexameters and Johnson's Irene

Tennyson's tragedies and Southey's epics—all these are classical

enough, serious enough, and literary enough, in all conscience. If

they entered the school library it would certainly be as Good Books,

not as Books. And it will hardly be denied that some of them are bad.

In fact, as soon as we look at the question from this point of view, it

becomes apparent that class A is not simply the class of bad books.

Mere failure does not infallibly give the right of entry to it. If all the

books in it are bad it must be with some special sort of badness—an A

badness.

But are they all bad? As I have not read the novels of Anthony

Hope, I cannot, though my third instance invites me, select them for

analysis; but perhaps Rider Haggard will do as well, for his books

are certainly in the A class and, in my opinion, some of them are

good—are therefore 'good "bad" books' as the Head ofthat college

would say. And of his books I select She. If I ask myself why it is

that I have more than once read She with enjoyment, I find that there

is every reason why I should have done so. In the first place the story

makes an excellent approach; the central theme is suffered, in the first

chapters, to woo us across great distances of space and time. What

we are presently to see at close quarters is seen at first, as it were,

through the wrong end of the telescope. This is a fine exercise in the

art of alluring—you may see the same thing at work in the, opening

of the Utopia, in the second act of Prometheus Unbound, and in the

early books of the Odyssey. In the second place it is a quest story,

which is an attractive thing. And the object of the quest combines

two strong appeals—it is the 'fountain of youth' theme and the

princesse lointaine in one. Finally, the withdrawal or conclusion,

1 Lewis later wrote An Experiment in Criticism (Cambridge, 1961) in which the 'experi-

ment' consists in judging literature itself by the way men read it. His purpose is, as he

says, ' to discover how far it might be plausible to define a good book as a book which

is read in one way, and a bad book as a book which is read in another' (p. 1).

268

HIGH AND LOW BROWS

which is always the difficulty in quest stories, is effected by unexpected

means which are nevertheless, on the author's suppositions, sufficiently

plausible. In the conduct of the story the detail is mostly convincing.

The characters who are meant to be amiable are amiable and those

who are meant to be sinister are sinister. The goodness of She is

grounded, as firmly as that of any book whatever, on the fundamental

laws of the imagination. But there is badness in it as well as goodness.

Two things deter us from regarding it as a quite satisfactory romance.

One of them is the continuous poverty of the style, by which, of

course, I do not mean any failure to conform to certain a priori rules,

but rather a sloth or incompetence of writing whereby the author is

content always with a vague approximation to the emotion, the

reflection, or the image he intends, so that a certain smudging and

banality is spread over all. The other fault is the shallowness and

folly of the things put into the mouth of She herself and offered us

for wisdom. That She, in her secular loneliness, should have become

a sage, is very proper, and indeed essential to Haggard's story, but

Haggard has not himself the wisdom wherewith to supply her. A

poet of Dante's depth could have given her things really wise to say;

a poet of Shakespeare's address would have made us believe in her

wisdom without committing himself.

If my analysis is correct, She is not a 'good "bad" book' in the

sense of being a good specimen of a bad kind; it is simply good and

bad, like many other books, in the sense that it is good in some

respects and bad in others. And those who have read it with enjoy-

ment have been enjoying real literary merits, and merits which it

shares with the Odyssey or The Life and Death of Jason. Certainly, it

is not a very good book, but since its vices are not sufficient to over-

whelm its virtues (as the experience of many wiser readers than I has

proved) it is better, say, than Leónidas or the Epic of Hades. In other

words, this book of the A class is better than some books of the B

class. It is better by every test; it shows more skill in the author and

produces more pleasure in the reader; it is more in touch with the

permanent nature of our imagination; it leaves those who have read

it richer. And with this, the attempt to identify the classes A and B

with the classes bad and good or worse and better has surely collapsed.1

1 For more detailed criticism of Haggard's romances see Lewis's essays, 'On Stories'

Of Other Worlds: Essays and Stories, ed. Walter Hooper (London, 1966), pp. 5 ff.

and 'Haggard Rides Again', Time and Tide, vol. xli (3 September i960), pp. 1044—5.

269

SELECTED LITERARY ESSAYS

We must now cast about for some other possible definition of the

two classes; and it comes at once into my mind that the lady in my

first instance ruled Peter Rabbit out as 'trivial'. Perhaps the A class

consists of trivial, and the B class of grave, or weighty, or momentous

books. I think that those who use the A—B antithesis very often have

something of this sort in their mind, but it is difficult to fix the exact

sense of any of the words they use to express it. It is clear that the

contrast cannot be simply between comic and serious, for then the

stories in our Parish Magazine would be labelled B and Le Misanthrope,

A. It might be argued, no doubt, that Molière's play, though comic,

is momentous as touching life at many points and dealing with the

depths of our nature, and this might furnish the ground for a restate-

ment of the distinction—B books being 'momentous' in the sense

just suggested, while A books touch us only superficially, or concern

only highly specialized areas of our consciousness. But this would

make The Importance of Being Earnest an A. Nor is it clear that all

the books already in the A class are thus superficial. They are often

accused of being ' sentimental ' ; and this charge, on inspection, often

conceals an admission that their appeal is to emotions very basic and

universal—to the same emotions that are concerned in great tragedies.

Even Dr Richards's Boosey Ballad,* which is not only A but bad

simply, has a theme that Petrarch might not have disdained. Indeed,

the more I look into it the more I am convinced that any contrast of

weighty and frivolous, solid and slight, deep and shallow, must cut

right across the A and B distinction. How many of the most perfect

things are, after all, trifles! The weighty and frivolous are kinds of

literature, and in each kind we shall find good A's, bad A's, good B's,

and bad B's. This is not what we are looking for.

Among the lowbrows themselves I find that the distinction is often

based on style. When the plain man confesses that the books he

delights in are not 'real Literature' he will often, if pressed, explain

this by saying that they 'haven't got style' or 'style and all that'.

And when the plain man has been captured and made into a pathe-

tically willing and bewildered university student he will sometimes

praise the great works which he has dutifully read and not enjoyed,

for the excellence of their style. He has missed the jokes in the comedy,

remained unmoved by the tragedy, failed to respond to the suggestions

of the lyric, and found the episodes of the romance uninteresting;

* See his Principles of Literary Criticism, ch. 24.

270

HIGH AND LOW BROWS

utterly at a loss to explain the value traditionally set on what has

proved to him so tedious, he hands it over to the thing he knows least

about, to a mysterious entity called Style, which is to him merely

what occult forces were to the old scientists—an asylum ignorantiae.

He does so because he has a radically false conception of style. He

thinks of it not as the linguistic means by which the writer produces

whatever results he desires but as a sort of extra—an uncovenanted

pedantry tacked on to the book proper, to gratify some specifically

'literary' or 'critical' taste which has nothing to do with the ordinary

pleasures of the imagination. It is for him a meaningless addition

which, by a convention, gives access to a higher rank—like the

letters Esq. after a man's name on an envelope. Now it must be

confessed that the highbrows sometimes talk of style in a way which

gives their weaker brethren some excuse for such misconceptions;

but I think that most of them, in a cool hour, would admit that

'Style', in the sense in which the lowbrow uses the word, does not

exist. When we say that the descriptions of country in She are marred

by their deficiency of style, we do not mean (as the ignorant suppose)

that they are good as descriptions but lacking in some abstractly

'literary' and undescriptive grace which might have been superadded;

we mean that they are imperfect descriptions; and we call their

imperfection 'stylistic' because it is due not to faults in the author's

conception but to his careless or insensitive language. A better choice

of epithets, and those distant mountains would have stood out sharper

on the horizon; a well chosen metaphor, and the whole picture, now

dimly discerned through seas of wasteful words, would have printed

itself for ever on the inner eye; a nobler rhythm, and the sense of

space and movement would have been given us, not left, as it now is,

for us to infer. Turn from She to almost any page in Eothen, and you

will soon see what style in a descriptive passage means. There is no

class of books which can be 'good in their own way' without bothering

about style. There are books in which what the author is saying

imperfectly, partly failing to say, is sufficiently interesting to keep us

reading in spite of his failure. Though he has done only half his work,

we are content to do some of it for him. Such books are books defective

in style. And they do not come only in class A. Scott, Dickens, Byron,

Spenser, Alanus, and Apuleius have detestable faults of style, but they

are usually put in class B. St Paul, despite some passages of striking

beauty, seems to me to write badly, but he is hardly an A; and I have

271

SELECTED LITERARY ESSAYS

found in the style of Donne, Chapman, Meredith, Saintsbury, and

others, obstacles to the enjoyment of what they have to give me as

great as those I find in Rider Haggard. It is not important here that any

reader should agree with the instances I am choosing; what matters

is the recognition that badness of style (like triviality) will be found

in B books. It cannot be made the basis of a dichotomy between two

kinds of book.

Another common way of using the distinction tends to fix on

'popular' as the best adjective for class A. 'Popular' art is supposed

to aim at mere entertainment, while 'real' or 'serious' art aims at

some specifically 'artistic' or 'aesthetic' or even 'spiritual' satis-

Th;

th

HIGH AND LOW BROWS

freak. What survives from most ages is chiefly either the work that

had some religious or national appeal, or else the popular, commercial

work produced for entertainment. I say 'chiefly' because the work of

the 'pure' artists is not always ephemeral; a little, a very little, of it

survives. But the great mass of literature which now fills class B is the

work of men who wrote either piously, to edify their fellows, or

commercially, to earn their living by 'giving the public what it

wanted'.

This leads to the very interesting conclusion that the B's of one

age have most often been the A's of another. We are sometimes

warned by the supporters of difficult new movements in literature

tint tn imitqfp rim

pre in crnt-iit-in- tha nennhoto