12 At the Fringe of Language
Language exists to communicate whatever it can communicate. Some things it communicates so badly that we never attempt to communicate them by words if any other medium is available. Those who think they are testing a boy’s ‘elementary’ command of English by asking him to describe in words how one ties one’s tie or what a pair of scissors is like, are far astray. For precisely what language can hardly do at all, and never does well, is to inform us about complex physical shapes and movements. Hence descriptions of such things in the ancient writers are nearly always unintelligible. Hence we never in real life voluntarily use language for this purpose; we draw a diagram or go through pantomimic gestures. The exercises which such examiners set are no more a test of ‘elementary’ linguistic competence than the most difficult bit of trick-riding from the circus ring is a test of elementary horsemanship.
Another grave limitation of language is that it cannot, like music or gesture, do more than one thing at once. However the words in a great poet’s phrase interinanimate one another and strike the mind as a quasi-instantaneous chord, yet, strictly speaking, each word must be read or heard before the next. That way, language is as unilinear as time. Hence, in narrative, the great difficulty of presenting a very complicated change which happens suddenly. If we do justice to the complexity, the time the reader must take over the passage will destroy the feeling of suddenness. If we get in the suddenness we shall not be able to get in the complexity. I am not saying that genius will not find its own ways of palliating this defect in the instrument; only that the instrument is in this way defective.
One of the most important and effective uses of language is the emotional. It is also, of course, wholly legitimate. We do not talk only in order to reason or to inform. We have to make love and quarrel, to propitiate and pardon, to rebuke, console, intercede, and arouse. ‘He that complains’, said Johnson, ‘acts like a man, like a social being.’ The real objection lies not against the language of emotion as such, but against language which, being in reality emotional, masquerades—whether by plain hypocrisy or subtler self-deceit—as being something else.
All my generation are much indebted to Dr I. A. Richards for having fully called our attention to the emotional functions of language. But I am hardly less indebted to Professor Empson for having pointed out that the conception of emotional language can be very easily extended too far. It was time to call a halt.
We must obviously not call any utterance ‘emotional’ language because it in fact arouses, even because it must arouse, emotion. ‘It is not cancer after all’, ‘The Germans have surrendered’, ‘I love you’—may all be true statements about matter of fact. And of course it is the facts, not the language, that arouse the emotion. In the last the fact communicated is itself the existence of an emotion but that makes no difference. Statements about crime are not criminal language; nor are statements about emotions necessarily emotional language. Nor, in my opinion, are value-judgements (‘this is good’, ‘this is bad’) emotional language. Approval and disapproval do not seem to me to be emotions. If we felt at all times about the things we judge good the emotion which is appropriate, our lives would be easier. It would also be an error to treat ‘I am washed in the blood of the Lamb’ as emotional language. It is of course metaphorical language. But by his metaphor the speaker is trying to communicate what he believes to be a fact. You may of course think the belief false in his particular case. You may think the real universe is such that no fact which corresponded to such a statement could possibly occur. You may say that the real cause which prompts a man to say things like that is a state of emotion. But if so, an emotion has produced erroneous belief about an impossible fact, and it is the fact erroneously believed in which the man is stating. A man’s hasty belief that the Germans had surrendered (before they did) might well be caused by his emotions. That would not make ‘The Germans have surrendered’ a specimen of emotional language. If you could find a man nowadays capable of believing, and saying, ‘The Russians have all been annihilated by magic’, even this would not be emotional language, though his belief in magic might be a belief engendered by emotion.
All this is fairly plain sailing. We reach something harder in the things said by poets. For there the purpose of the utterance would be frustrated if no emotion were aroused. They do not merely, like the sentences cited above, arouse emotion in fact; it is their purpose—at any rate, part of their purpose—to do so. But we must be very careful here. Having observed that a poetical utterance in fact arouses emotion, and is intended to arouse emotion, and that if taken as a statement about reality—or even about the make-believe ‘realities’ of a fictitious narrative—it would be nonsensical or at least false, can we conclude that it communicates nothing but emotion? I think not.
Nothing will convince me that ‘My soul is an enchanted boat’ is simply a better way—however much better—of doing what might be done by some exclamation like ‘Gee!’ Asia has risen from the dark cave of Demogorgon. She is floating upwards. She is saluted as ‘Life of Life!’ The reversed temporal process in ll. 97–103 (‘We have passed Age’s icy caves’ etc.), borrowed from Plato’s Politicus (269c sq.), marks the fact that at this moment the whole cycle is reversed and cosmos begins anew. She is undergoing apotheosis. What did it feel like? The poet says to us in effect ‘Think of going in a boat. But quite effortless’ (‘Like a sleeping swan’ gliding with the current, he adds in the next line), ‘Like a boat without sail or oar; the motive power undiscoverable. Like a magic boat—you must have read or dreamed of such things—a boat drawn on, drawn swiftly on, irresistibly, smoothly, by enchantment.’ Exactly. I know now how it felt for Asia. The phrase has communicated emotion. But notice how. By addressing in the first instance my imagination. He makes me imagine a boat rushing over waves, which are also identified with sounds. After that he need do no more; my emotion will follow of itself. Poetry most often communicates emotions, not directly, but by creating imaginatively the grounds for those emotions. It therefore communicates something more than emotion; only by means of that something more does it communicate the emotion at all.
Burns compares his mistress to ‘a red, red rose’; Wordsworth his to ‘a violet by a mossy stone Half hidden from the eye’. These expressions do communicate to me the emotion each poet felt. But it seems to me that they do so solely by forcing me to imagine two (very different) women. I see the rose-like, overpowering, midsummer sweetness of the one; the reticent, elusive freshness, the beauty easily overlooked in the other. After that my emotions may be left to themselves. The poets have done their part.
This, which is eminently true of poetry, is true of all imaginative writing. One of the first things we have to say to a beginner who has brought us his MS. is, ‘Avoid all epithets which are merely emotional. It is no use telling us that something was “mysterious” or “loathsome” or “awe-inspiring” or “voluptuous”. Do you think your readers will believe you just because you say so? You must go quite a different way to work. By direct description, by metaphor and simile, by secretly evoking powerful associations, by offering the right stimuli to our nerves (in the right degree and the right order), and by the very beat and vowel-melody and length and brevity of your sentences, you must bring it about that we, we readers, not you, exclaim “how mysterious!” or “loathsome” or whatever it is. Let me taste for myself, and you’ll have no need to tell me how I should react to the flavour.’
In Donne’s couplet
Your gown going off, such beautious state reveals
As when from flowry meads th’hills shadow steales
beautious is the only word of the whole seventeen which is doing no work.
There are exceptions to this principle. By very successful placing, a great author may sometimes raise such words to poetic life. Wordsworth’s lines are a specimen:
Which, to the boundaries of space and time,
Of melancholy space and doleful time,
Superior—
Here we have almost the reverse of the process I have been describing. The object (space and time) is in one way so familiar to our imaginations and in another so unimaginable—we have read so many tedious attempts to exalt or over-awe us with mere superlatives or even with simple arithmetic—that nothing can be made of it. This time, therefore, the poet withdraws the object (the ground for emotion) altogether and appeals directly to our emotions; and not to the quite obvious ones. Another exception is naturally to be found in drama or very dramatic lyric, where the poet—with discretion and a proper use of illusion—imitates the speech of people in some highly emotional situation—even, at need, their inarticulate cries. This in its purity, which purity a good poet never sustains for long, belongs to poetry not in so far as poetry is a special use of language but in so far as poetry is mimesis. In themselves the ‘Ah! Ah!’ or ‘Otototoi’ or ‘Iou! Iou!’ of characters in a Greek tragedy are not specimens of poetry any more than the ‘Bé, bé’ of the lamb or the ‘Au! Au!’ of the dog in Aristophanes.
In general, however, the poet’s route to our emotions lies through our imaginations.
We must also exclude from the category ‘emotional language’ words such as I have taken supernatural to be. The class of things which they refer to may be bound together chiefly by a common emotion; but the purpose of using the words is to assign something to that class, not merely to communicate the emotion which led to the classification.
Having thus narrowed the field, we can now make a new start. It will be noticed that I have throughout used the word emotional rather than emotive. This is because I think the latter word applicable to only one aspect of emotional language. For an ‘emotive word’ ought to mean one whose function is to arouse emotion. But surely we ought to distinguish utterances which arouse, from those which express, emotion? The first is directed towards producing some effect on a (real or imagined) hearer; the second discharges our own emotion, cleanses our stuffed bosom of some perilous stuff.
The distinction will seem straw-splitting if we have in mind the language of love. For, as Samson says, ‘love seeks to have love’, and it would be hard to say whether endearments serve more as expressions of love in the speaker or incitements to it in the beloved. But that tells us more about the nature of love than about the nature of language. One of my old headmasters once wisely said it was a pity that amare was the first Latin verb we all learn. He thought this led to an imperfect grasp of the difference between the active and the passive voice. It might be better to begin with flagellare. The difference between flogging and being flogged would come home to the business and bosoms of schoolboys far more effectively than that of loving and being loved. On the same principle, we can best see the distinction between the stimulant and the expressive functions of emotional language in a quarrel; and best of all where the same word performs both. The man who calls me a low hound both expresses and (actually or intentionally) stimulates emotion. But not the same emotion. He expresses contempt; he stimulates, or hopes to stimulate, the almost opposite emotion of humiliation.
Again, in the language of complaint we often find the expressive without the stimulant. When two people who have missed the last train stand on the silent platform saying ‘Damn’ or ‘Bloody’ or ‘Sickening’, they neither intend nor need to stimulate each other’s disappointment. They are just ‘getting it off their chests’.
The vocabulary of endearment, complaint, and abuse, provides, I think, almost the only specimens of words that are purely emotional, words from which all imaginative or conceptual content has vanished, so that they have no function at all but to express or stimulate emotion, or both. And an examination of them soon convinces us that in them we see language at its least linguistic. We have come to the frontier between language and inarticulate vocal sounds. And at that frontier we find a two-way traffic going on.
On the one hand we find inarticulate sounds becoming words with a fixed spelling and a niche in the dictionary. Thus English heigh-ho and Latin eheu are clearly formalised imitations of the sigh; ah, of the gasp; tut-tut, of the tongue clicked against the hard palate. These are general. In particular situations the ‘verbification’ of the inarticulate may occur ad hoc. A voluntary scream may become a cry for mercy. A voluntary groan, from a wounded man, uttered to attract the attention of the stretcher-bearers, may be the equivalent of a sentence (‘There is a wounded man in this ditch’).
But we also see the frontier being crossed in the opposite direction. In the vocabulary of abuse and complaint we see things that once were words passing out of the realm of language (properly so called) and becoming the equivalents of inarticulate sounds or even of actions; of sighs, moans, whimperings, growls, or blows.
The ‘swear-words’—damn for complaint and damn you for abuse—are a good example. Historically the whole Christian eschatology lies behind them. If no one had ever consigned his enemy to the eternal fires and believed that there were eternal fires to receive him, these ejaculations would never have existed. But inflation, the spontaneous hyperboles of ill temper, and the decay of religion, have long since emptied them of that lurid content. Those who have no belief in damnation—and some who have—now damn inanimate objects which would on any view be ineligible for it. The word is no longer an imprecation. It is hardly, in the full sense, a word at all when so used. Its popularity probably owes as much to its resounding phonetic virtues as to any, even fanciful, association with hell. It has ceased to be profane. It has also become very much less forceful. You may say the same of sickening in its popular, ejaculatory, use. There are alarms and disappointments which can actually produce nausea, or, at least, emotions which we feel to be somehow similar to it. But the man who says sickening! when he has missed the train is not thinking about that. The word is simply an alternative to damn or bloody. And of course far weaker than it would be if it still carried any suggestion of vomiting.
So with abusive terms. No one would now call his schoolfellow or next door neighbour a swine unless someone had once used this word to make a real comparison between his enemy and a pig. It is now a mere alternative to beast or brute or various popular unprintable words. They are all interchangeable. Villain, as we know, once really compared your enemy to a villein. Once, to call a man cad or knave assigned to him the status of a servant. And it did so because, earlier still, these words meant ‘boy’ or ‘junior’ (you address a slave as ‘boy’ in Greek and a waiter as garçon in French).
Thus all these words have come down in the world. None of them started by being merely abusive, few of them by being abusive at all. They once stimulated emotion by suggesting an image. They made the enemy odious or contemptible by asserting he was like somebody or something we already disliked or looked down on. Their use was a sort of passionate parody of the syllogism: pigs (or servants or my juniors) are contemptible—John is like a pig (or servant or adolescent)—therefore John is contemptible. That was why they really hurt; because hurting was not the whole of what they did. They stimulated emotion because they also stimulated something else; imagination. They stimulated emotion in the particular case because they exploited emotions which already existed towards whole classes of things or persons. Now that they are nothing whatever but emotional stimulants, they are weak emotional stimulants. They make no particular accusation. They tell us nothing except that the speaker has lost his temper.
And even this they do not tell us linguistically, but symptomatically; as a red face, a loud voice, or a clenched fist, might do equally well. The fact of the other person’s anger may hurt or frighten us; hurt us if we love him, or frighten us if he is larger and younger than ourselves and threatens violence. But his language as such has very little power to do the only thing it is intended to do. It would have been far more wounding to be called swine when the word still carried some whiff of the sty and some echo of a grunt; far more wounding to be called a villain when this still conjured up an image of the unwashed, malodorous, ineducable, gross, belching, close-fisted, and surly boor. Now, who cares? Language meant solely to hurt hurts strangely little.
This can be seen clearly when we catch a word ‘just on the turn’. Bitch is one. Till recently—and still in the proper contexts—this accused a woman of one particular fault and appealed, with some success, to our contempt by calling up an image of the she-dog’s comical and indecorous behaviour when she is on heat. But it is now increasingly used of any woman whom the speaker, for whatever reason, is annoyed with—the female driver who is in front of him, or a female magistrate whom he thinks unjust. Clearly, the word is far more wounding in its narrower usage. If that usage is ever totally lost—as I think it will be—the word will sink to the level of damn her. Notice, too, how cat (of a woman) is still strong and useful because the image is still alive in it.
An important principle thus emerges. In general, emotional words, to be effective, must not be solely emotional. What expresses or stimulates emotion directly, without the intervention of an image or concept, expresses or stimulates it feebly. And in particular, when words of abuse have hurting the enemy as their direct and only object, they do not hurt him much. In the field of language, however it may be in that of action, hatred cuts its own throat, and those who are too ‘willing to wound’ become thereby impotent to strike. And all this is only another way of saying that as words become exclusively emotional they cease to be words and therefore of course cease to perform any strictly linguistic function. They operate as growls or barks or tears. ‘Exclusively’ is an important adverb here. They die as words not because there is too much emotion in them but because there is too little—and finally nothing at all—of anything else.
In this there is not much to be lamented. If a mother with a baby, or lovers in each other’s arms, use language so emotional that it is really not language at all, I see no ground for shame or offence; and if men in an orgy of resentment, though (in the physical sense) they articulate, are really no more speaking—are saying no more—than a snarling animal, this is perhaps all for the best. The real corruption comes when men whose purpose in speaking is in fact purely emotional conceal this from others, and perhaps from themselves, by words that seem to be, but are not, charged with a conceptual content.
We have all heard bolshevist, fascist, Jew, and capitalist, used not to describe but merely to insult. Rose Macaulay noticed a tendency to prefix ‘so called’ to almost any adjective when it was used of those the speaker hated; the final absurdity being reached when people referred to the Germans as ‘these so-called Germans’. Bourgeois and middle class often suffer the same fate.
A literary man of my acquaintance, on reading an unfavourable reference to his own works, called it vulgar. The charge brought against him was one that only highly educated people ever bring; the tone of the passage not otherwise offensive than by being unfavourable; the phrasing perfectly good English. If he had called it false, unintelligent, or malicious, I could have understood, though I might have disagreed. But why vulgar? Clearly, this word was selected solely because the speaker thought it was the one that the enemy, if he could hear it, would most dislike. It was the equivalent of an oath or a growl. But that was concealed from the speaker because ‘This is vulgar’ sounds like a judgement.
When we write criticism we have to be continually on our guard against this sort of thing. If we honestly believe a work to be very bad we cannot help hating it. The function of criticism, however, is ‘to get ourselves out of the way and let humanity decide’; not to discharge our hatred but to expose the grounds for it; not to vilify faults but to diagnose and exhibit them. Unfortunately to express our hatred and to revenge ourselves is easier and more agreeable. Hence there is a tendency to select our pejorative epithets with a view not to their accuracy but to their power of hurting. If writing which was intended to be comic has set our teeth on edge, how easily the adjectives arch or facetious trickle out of the pen! But if we do not know exactly what we mean by them, if we are not prepared to say how comic work which errs by archness and facetiousness differs from comic work which errs in any other way, it is to be feared that we are really using them not to inform the reader but to annoy the author—arch or facetious being among the most effective ‘smear-words’ of our period. In the same way work which obviously aspires and claims to be mature, if the critic dislikes it, will be called adolescent; not because the critic has really seen that its faults are those of adolescence but because he has seen that adolescence is the last thing the author wishes or expects to be accused of.
The best protection against this is to remind ourselves again and again what the proper function of pejorative words is. The ultimate, simplest and most abstract, is bad itself. The only good purpose for ever departing from that monosyllable when we condemn anything is to be more specific, to answer the question ‘Bad in what way?’ Pejorative words are rightly used only when they do this. Swine, as a term of abuse is now a bad pejorative word, because it brings no one accusation rather than another against the person it vilifies; coward and liar are good ones because they charge a man with a particular fault—of which he might be proved guilty or innocent. As applied to literature, dull, hackneyed, incoherent, monotonous, pornographic, cacophonous, are good pejoratives; they tell people in what particular way we think a book faulty. Adolescent or provincial are not so good. For even when they are honestly used, to define, not merely to hurt, they really suggest a cause for the book’s badness instead of describing the badness itself. We are saying in effect ‘He was led into his faults by being immature’ or ‘by living in Lancashire’. But would it not be more interesting to indicate the faults themselves and leave out our historical theory about their causes? If we find words like these—and vulgar, and others—indispensable to our criticism, if we find ourselves applying them to more and more different kinds of things, there is grave reason to suspect that—whether we know it or not—we are really using them not to diagnose but to hurt. If so, we are assisting in verbicide. For this is the downward path which leads to the graveyard of murdered words. First they are purely descriptive; adolescent tells us a man’s age, villain, his status. Then they are specifically pejorative; adolescent tells us that a man’s work displays ‘mawkishness and all the thousand bitters’ confessed by Keats, and villain tells that a man has a churl’s mind and manners. Then they become mere pejoratives, useless synonyms for bad, as villain did and as adolescent may do if we aren’t careful. Finally they become terms of abuse and cease to be language in the full sense at all.
As this book is now almost done, what would otherwise be a digression—for it carries us beyond the subject of vocabulary—may perhaps be excused as a sort of coda. In the last few paragraphs we have had to touch on criticism. I would be very glad if I could transfer to even one reader my conviction that adverse criticism, far from being the easiest, is one of the hardest things in the world to do well. And that for two reasons.
Dr I. A. Richards first seriously raised the problem of badness in literature. And his singularly honest wrestling with it shows how dark a problem it is. For when we try to define the badness of a work, we usually end by calling it bad on the strength of characteristics which we can find also in good work. Dr Richards began by hoping he had found the secret of badness in an appeal to stock responses. But Gray’s Elegy beat him. Here was a good poem which made that appeal throughout. Worse still, its particular goodness depended on doing so. This happens again and again. The novel before you is bad—a transparent compensatory fantasy projected by a poor, plain woman, erotically starving. Yes, but so is Jane Eyre. Another bad book is amorphous; but so is Tristram Shandy. An author betrays shocking indifference to all the great political, social, and intellectual upheavals of his age; like Jane Austen. The solution of the problem is, I suspect, still far away.
The other difficulty lies within. As I said before, what we think thoroughly bad, we hate. If, besides being bad, it enjoys great popularity and thereby helps to exclude works that we approve from their ‘place in the sun’, hatred of a somewhat less disinterested sort will creep in. Lower and still lower levels of hatred may open; we may dislike the author personally, he and we may belong to opposed literary ‘parties’ or factions. The book before us becomes a symbol of l’infâme. Hence a perpetual danger of what is called criticism (judgement) becoming mere action—a blow delivered in a battle. But if it does, we are lost as critics.
Everyone who remembers Arnold’s ‘Literary Influence of Academies’ will see why we are lost. But its lesson has been forgotten. There has been in our time a determined, and successful, attempt to revive the brutalité des journaux anglais. Reviews so filled with venom have often been condemned socially for their bad manners, or ethically for their spite. I am not prepared to defend them from either charge; but I prefer to stress their inutility.
They can, no doubt, be enjoyed if we already agree with the critic. But then, you know, we are not reading them to inform our judgement. What we enjoy is a resounding blow by our own ‘side’. How useless they are for any strictly critical function becomes apparent if we approach them with an open mind. I had this forced upon me when I read some unusually violent reviews lately which were all by the same man. My mind could not but be open. The books he reviewed were not by me nor by any close friend of mine. I had never heard of the critic. I read (at first—one soon learned to skip his productions) to find out what the books were like and whether I should consider buying them. But I found I could learn nothing about the books. In the first hundred words the critic had revealed his passions. What happened to me after that is, I think, what must happen to anyone in such circumstances. Automatically, without thinking about it, willy-nilly, one’s mind discounts everything he says; as it does when we are listening to a drunk or delirious man. Indeed we cannot even think about the book under discussion. The critic rivets our attention on himself. The spectacle of a man thus writhing in the mixed smart and titillation of a fully indulged resentment is, in its way, too big a thing to leave us free for any literary considerations. We are in the presence of tragicomedy from real life. When we get to the end we find that the critic has told us everything about himself and nothing about the book.
Thus in criticism, as in vocabulary, hatred over-reaches itself. Willingness to wound, too intense and naked, becomes impotent to do the desired mischief.
Of course, if we are to be critics, we must condemn as well as praise; we must sometimes condemn totally and severely. But we must obviously be very careful; in their condemnations great critics long before our time have exposed themselves. Is there any way in which we—lesser men than they—can avoid doing the same? I think perhaps there is. I think we must get it firmly fixed in our minds that the very occasions on which we should most like to write a slashing review are precisely those on which we had much better hold our tongues. The very desire is a danger signal. When an author whom we admire in general, writing in a genre we thoroughly enjoy, produces a disappointing work, we may proceed with tolerable safety. We know what we had hoped for. We see, and would have relished, what he was trying to do. By that light we may possibly diagnose where the book has gone wrong. But when an author we never could stand is attempting (unsuccessfully—or, worse still, successfully) ‘exactly the sort of thing we always loathe’, then, if we are wise, we shall be silent. The strength of our dislike is itself a probable symptom that all is not well within; that some raw place in our psychology has been touched, or else that some personal or partisan motive is secretly at work. If we were simply exercising judgement we should be calmer; less anxious to speak. And if we do speak, we shall almost certainly make fools of ourselves.
Continence in this matter is no doubt painful. But, after all, you can always write your slashing review now and drop it into the wastepaper basket a day or so later. A few re-readings in cold blood will often make this quite easy.