Jottings Logo - John Fraser

Voices in the Cave of Being

Winters, Leavis, and Language


As William Booth, the founder of the Salvation Army, said about its brassy music, “Why should the Devil have all the good tunes?”

Meaning, here, what chutzpah it is to imply that “theory,” literary theory, didn’t really begin until Derrida and De Man, in their twin act, hit East Coast American academe like the Beatles.

I am thinking, especially, of the idea that language needs to be “problematized,” in particular the presumption of a one-to-one correspondence between words, and, for want of a better word, things.

Derrida was obviously engaged in a long, susperstitious, and, at bottom, unsuccessful struggle to free himself from the twin claims of Platonism and the Kabbalah with respect to the intrinsic truth-bearing nature of language.

De Man, equally obviously, was operating within the framework of a naïve Cartesian dualism, with, on the one hand, the impersonal and inorganic real universe of physics (there’s a sentence of his somewhere or other where he actually mentions “atoms”!) and, on the other, the unanchored, arbitrary games of language—of mere words and, insofar as words are used to articulate values, “mere” values.

I won’t be mentioning either of them again here.


Early in Yvor Winters’In Defencs of Reason there’s an exemplary passage from 1937 about the nature of poetry. I have broken up the paragraphs and provided numbers to make reading and commenting easier. The emphases are mine.


(1) [L]anguage is a kind of abstraction, even at its most concrete; such a word as “cat,” for instance, is generic and not particular. Such a word becomes particular only in so far as it gets into some kind of experiential complex, which qualifies it and limits it, which gives it, in short, a local habitation as well as a name.

(As in, let’s say, Thomas Gray’s “On a Favourite Cat, Drowned in a Tub of Gold Fishes,” or William Carlos Williams’ “Poem,” or Stevie Smith’s “The Galloping Cat,” or the yellow fog passage in T.S. Eliot’s,“The Love Song of J. Alfred Prufrock,” or Christopher Smart’s “For I will consider my cat Jeoffrey,” or B. Kliban’s “Love to eat them mousies,” or Eliot’s “The Naming of Cats,” or Gavin Ewart’s “A 14-year Old Convalescent Cat in the Winter.”)

(2) Such a complex is the poetic line or other unit, which, in turn, should be a functioning part of the larger complex, or poem.

[No cat is instantly there (the sign “full”) in a poem. A “cat”, or cattishness, enters and starts becoming, phrase by phrase, line by line, like Ted Hughes’ “Thought-Fox,” where first there’s just a nose “touching twig, leaf,” and then the movement of wary eyes, and footprints appearing in the snow, and finally “a sudden sharp hot stink of fox.”

A diversity of felines: “I want him to have another living summer,” “The naming of cats is a difficult matter,” “I am a cat that likes to/ Gallop about doing good,” “As the cat/ climbed over/the top of,” “The yellow fog that rubs its muzzle on the window panes,” “’Twas on a lofty vase’s side,” “Love to eat them mousies” (“Mousies what I love to eat,/ Bite they little heads off…/ Nibble on they tiny feet”).

The signifier of straight lines and curves c/a/t is empty. As are c/ha/t/, k/a/t/z, g/a/t/o/, and § (my own ad hoc signifier today).

They are like small circles, some solid, some hollow, set on and between thin horizontal lines, meaningless unless “heard,” whether spectrally for your own voice, or a cello, or xylophones, or massed choirs. (They are meaningless to me. I “hear” nothing at all, beyond the vague recognition that sounds must be going up or down in pitch)

The sound “katt” (or “pussy”) that a child makes is always a pointing—a house pet walking, or sleeping, or acting up, an unfamiliar tom strolling across the summer lawn, a cat in a hat dancing in a book—those shapes, those movements, all part (or parts) of the child’s “sign.”

The words of mists-of-time ancestral communities that we’d translate as “tiger” weren’t just said, like coughs or hiccups. They came with, were part of, that particular dead animal, or prowling live one, or suspicious noise in the thickets, or those things being recalled in narratives.

All of which gets pared away later during the production of essentialized definitions for legal or scientific purposes.

I still remember being confronted in a scholarship exam with, “Define an orange.”

Impossible! For me, anyway.

An orange was its pitted orange rind, and the pith your finger nails started digging into as you burrowed through it, and the segments in their membranes, and the tingling juice that spurted into your mouth when you bit into one of them.

It was nice coming a dozen years ago on Rilke’s advice/command, as translated by David Young (Tanzt die Orange/ Dance the orange), to

Peel away, radiant,
Fragrance on fragrance! Create a kinship
With the pure and reluctant rind,
With the juice that loads the ecstatic fruit!.]

(2) This is, I imagine, what Mallarmé should have had in mind when he demanded that the poetic line be a new word, not found in any dictionary, and partaking of the nature of incantations (that is, having the power to materialize, or perhaps it would be more accurate to say, being a new experience).

[“Hilare or de cymbale à des poings irrité” (Mallarmé)—hilarious cymbal gold angered/excited by fists”—and no, this doesn’t simply dwindle down to the one word “sun.” It’s a new seeing/feeling/experiencing of that shape in the sky, that illumination, that warmth on the skin.]

(3) The poem, to be perfect, should likewise be a new word in the same sense, a word of which the line, as we have defined it, is merely a syllable.

[Leavis spoke approvingly of a 1936 review-article on T.S. Eliot in which D.W. Harding suggested that

one could say, perhaps, that [Burnt Norton] takes the place of the ideas of ‘regret’ and ‘eternity.’ Where in ordinary speech we should have to use those words, and hope by conversational trial-and-error to obviate the grossest misunderstandings, this poem is a newly-created concept, equally abstract but vastly more exact and rich in meaning.”]

(4) Such a word is, of course, composed of much more than the sum of its words (as one normally uses the term) and its syntax.

(5) It is composed of an almost fluid complex, if the adjective and the noun are not too nearly contradictory, of relationships between words (in the normal sense of the term), a relationship involving rational content, cadences, rhymes, juxtapositions, literary and other connotations, inversions, and so on, almost indefinitely.

[Yes indeed, though not all necessarily in the same poem.]

(6) These relationships, it should be obvious, extend the poet’s vocabulary incalculably. They partake of the fluidity and unpredictability of experience and so provide a means of treating experience with precision and freedom.


(7) If the poet does not wish, as actually he seldom does, to reproduce a given experience with approximate exactitude, he can employ the experience as a basis for a new experience that will be just as real, in the sense of being particular, and perhaps more valuable.

[The poet might remember a summer night when he was in the fields going to meet his lover and there were some glowworms, and he could describe that. Or he could be Andrew Marvell and write “The Mower to the Glow-Worms.”]

(8) Now verse is more valuable than prose in this process for the simple reason that its rhythms are faster and more highly organized than are those of prose, and so lend themselves to a greater complexity and compression of relationship, and that the intensity of this convention renders possible a greater intensity of other desirable conventions, such as poetic language and devices of rhetoric.

[The term “prose” is a much broader-spectrum one than “poetry.” But there are rhythmic phrasings of some kind, clunky or elaborate, plangent or staccato, etcera, in passages of prose, and it’s possible to use most of the other resources that Winters lists, so that in fact there is no sharp boundary between “prose” and “poetry.” It’s a spectrum, not a dichotomy. If you read out loud a passage of “verse” and a passage of “prose,” you are both times reading out sequences of phrasal units.]

(9) The writer of prose must substitute bulk for this kind of intensity; he must define his experience ordinarily by giving all of its past history, the narrative logic leading up to it, whereas the experiential relations given in a good lyric poem, though particular in themselves, are applicable without alteration to a good many past histories.

[You’d look very odd announcing out of the blue on a sheet of paper—and where?—that someone you were in love with reminded you of a rose and nice tunes and that you’d always love her dearly. Who is this man, who’s this woman, how old are they, and so forth? A fiction writer or dramatist would have to do quite a bit of filling in.

But if you say,

O my Luve’s like a red red rose
That’s newly sprung in June;
O my Luve’s like the melodie
That’s sweetly played in tune,

and the reader, at least one who’s read other poems, sees that the lines don’t go all the way to the right margin and that there’s a title at the top, he or she won’t be bothered.

All manner of individuals, young, old, rich, poor, English, Asian-Indian, etc, can feel that Burns’ poem describes how they themselves feel—provides a concept that more accurately fits their feelings than simply saying “I’m in love. I’m really in love. I’ll love her forever” would do. The phrase “in love” simply doesn’t do the work (“Please answer the question. Yes or no, did you or did you not love the plaintiff?”)

W.H. Auden has a charming poem, “O Tell Me the Truth about Love.” (“Does it look like a pair of pyjamas/Or the ham in a temperance hotel?” and so on.)]

(10) In this sense, the lyric is general as well as particular; in fact, this quality of transferable or generalized experience might be regarded as the defining quality of lyrical poetry.

[Which doesn’t mean that it always happens, or that a poem can’t be considered a lyric poem if it doesn’t happen, but that it can happen—and happen faster than in a play or novel, where it takes a while to “receive” what’s going on.]

(11)What I have just said should make plain the difficulty of comprehending a poem exactly and fully; its total intention may be very different from its paraphrasable, or purely logical contents.

[Emphasis mine. Maybe this is the element of truth that is obfuscated in portentous statements about the impossibility of “understanding” a literary text, if by “understanding” is meant the ability to give a totalizing account that would reproduce the effect of (and virtually be) the original text.]


Linguistically, what Winters has said in this passage is pretty sophisticated with respect to both the alleged indeterminacy of words and the alleged claim that words only “mean” in relation to other words.

(He was in his thirties and reading Mallarmé intently at a time when Clever Jacques was still in knee pants.)

If a rural Texan tells you that a friend of his is “a good ol’ boy,” he’s not saying that he’s good (in the conventional moral sense) and old and a boy. The friend may not be any of those things. The phrase “ good old boy,” with its definite Southern meaning, is in effect a new word, or term, or, if you prefer, sign.

And the three terms that have been dissolved into it are themselves not all that definite to start with. It is really only convenience that puts “old” in the sense of “advanced in years” and “old” in the sense of “former” in the same dictionary slot. The signifiers may look the same, the way a couple of matches do, but the signifieds are sharply different and only come into being in context.

The phrases “my old cat” and “my old chemistry teacher” (as distinct from “my old chemistry teacher,” meaning not my young one), are in effect different signs, or terms, or, if you prefer, “words.”

But with a modicum of familiarity with Southern speech, you would still know that if you were to call on the neighbour you wouldn’t find a scrawny, pinch-mouthed, Bible-quoting moralist who wouldn’t offer you a drink.

Some of us didn’t have to wait for Derrida and De Man to liberate us from naïve one-to-one-correspondence ideas of language.


The point of all this, I mean why I’m presenting Winters’ passage here (over and above offering a reminder that my favourite critic of poetry did not, repeat not, say that poems must always be offering rational arguments), is that it’s the repertoire of local precisions and at times almost invisible but cumulative metrical adjustments that make possible the expressive precision of the whole work.

Robert Burns didn’t try to validate his love, or the speaker’s love, by claiming the authority of autobiography (“I’m a very loving person with a deep capacity for love, as I’ve noticed with pleasure over the years and been told about by friends, so that when I say I love here you can be quite sure that I really do love her, because I know myself very well, the way my favourite poet William Wordsworth knew himself.”

Or by pointing to some extra-personal force or presence. (“The Divine Spirit of Love which animates the Universe and is at work in us all has taken possession of me and allowed no room for any earthly doubts, so that I can solemnly affirm with my hand pressed to my heart that I love her totally and beyond words, as used to happen also to my favourite poet P.B. Shelley.”

Either, you might be inclined to think, what a jerk! (quel con!) Or wonder what Dickens might have done with such a character.

Instead, after the opening rose-and-sweet-melody similes, Burns gives us:

As fair art thou, my bonnie lass,
So deep in luve am I;
And I will love thee still, my Dear,
Till a’ the seas gang dry.—

Till a’ the seas gang dry, my Dear,
And the rocks melt wi’ the sun;
I will love thee still, my Dear,
While the sands o’ life shall run.—.


If you don’t get some kind of tingle from the combining of the colloquial “bonnie lass” and “my Dear” with the lift-off into vast time and space with all the seas going dry (and “deep” contributing an initial sense of wetness)

—and then the intensification of that with the repetition of the line

—and the tactile extension of the image with the vast heat of the sun melting the exposed rocks (he hadn’t just casually grabbed a passing metaphor)

—and the intensification there with the reversed foot that throws an unexpected stress on “melt”

—and then the return from mega-vistas to the more domesticated image of the hour-glass , but with a hint of rocks not only melted but dried to sand, and the contrasting memory trace of the fresh spring flower in the opening stanza;

And if someone were to object that the speaker isn’t going to be able to live that long, so he’s a manipulative braggart; or that we can’t tell whether he’s being sincere or not unless we know the circumstances of the poem’s composition—well, poetry might not really be their cup of tea, might it?

Personally I don’t need to know more than the poem itself, any more than did and do, I imagine, the tens, perhaps hundreds of thousands of. readers who have loved it.

It would be no less effective were it anonymous or virtually anonymous, like Mark Alexander Boyd’s “Fra bank to bank.”

And you can see how, yes, it could indeed function like a “word” or sign (in Saussure’s sense), so that instead of saying, “I’m in love,” one could say (or think), “I feel [like] that poem.”.


In a brief statement designed to help get Ezra Pound released from St. Elizabeth’s mental hospital, Hemingway said that it was no more possible not to have been influenced as a writer by Pound than to traverse a desert without feeling the heat.

What F.R. Leavis says about figurative language in a few pages of Education and the University (1943) must have imprinted similarly on a lot of readers.

Leavis moves in on Matthew Arnold’s sonnet “To Shakespeare,” the one that begins, “”Others abide our question. Thou art free./ We ask and ask. Thou smilest and art still,” and pounces on the comparison of Shakespeare to a lofty hill that you see “Planting his steadfast footsteps in the sea.” It introduces, as he rightly says, “a ludicrous suggestion of gigantic, ponderously wading strides.”

(Three years later, in “Politics and the English Language,” George Orwell would give us the Fascist octopus singing its many-armed swan-song.)


So (polite yawn behind back of hand), mixed tropes, eh?

Well, not quite, or least that’s not all. After observing that Arnold’s phrasing “could only have been offered by an unrealizing mind, handling words from the outside,” Leavis goes on to remark that:

But it will not do to say simply that in good poetry the metaphors are realized. In fact there are hardly any rules that can, with any profit, be laid down: the best critical terms and concepts one can find or provide oneself with will be inadequate to the varied complexities with which the critic has to deal.

And he offers in half a dozen pages some superb brief analyses of several passages from Macbeth of which you can indeed feel that words are being used from “inside” Macbeth soliloquizing just before Duncan’s murder (the murder of a king) in terms that go beyond (or not as far as) a merely visual comparison of one definite thing with another.

Quoting the sentence

I have no spur
To prick the sides of my intent, but only
Vaulting ambition, which o’erleaps itself
And falls on the other [side],

he comments:

Macbeth’s ‘intent’ of murder is, to his feeling, quite other than himself; as external to himself as an unwilling horse between his thighs; he can muster no impulse sharp enough to prick it into action. Then, with a rapid change in his psychological relation to the horse, he expresses the sense of difficulty and danger that produces this paralysis—a sense at the same time of the supreme effort required (he is gathered tense for vaulting—this point in the speech is a good instance of the expressive use of line division) and of the terrifying impossibility of making sure that the process once started can be stopped at the point of achievement in view.

And then comes a further caution: “It will not do to treat metaphors, images and other local effects as if their relation to the poem were at all like that of plums to cake, or stones attesting that the jam is genuine.”


If Leavis hadn’t “attacked” Milton and Shelley, his reputation would surely have been a good deal different.

His deflationary remarks about Paradise Lost infuriated the kinds of readers who wanted to feel that here was, if not the voice of God, at least a Godlike voice that you could be borne along by trustingly.

But worse, I suspect, was Leavis’s examination of the figurative language in Shelley’s “When the Lamp is Shattered,” his scrutinizing of the metaphors, his testing of them out in truth-to terms (light lying in the dust after a lamp is broken?), his teasing out of the syntax to figure out exactly who is being commiserated with.

You can sense, as you read the analysis now, the resentment of readers back then who just knew how the young-god-like Shelley felt, and who believed themselves that the world (and Love) were indeed as he seemed to be saying they were, and who sensed that they themselves were being brought into question in the analysis, and that that was not what they went to poetry for.

Poetry was supposed to be a simple fit—the poetic self and How Things Really Were.

Here, instead, was Leavis focusing on the heuristic process as word followed word, and engaging in a species of deconstruction long before Paul de Man, infuriated by the New York Review of Books’ rejection of one of his smoothly literary-journalistic review-articles, swerved away to the kind of intent pouncing on “betraying” details (often tropes), and refusal to let things drop in argument as if at the end of a well-played set of tennis, that Leavis had displayed in Scrutiny in his 1935 review-article on I.A. Richards’ Coleridge on Imagination and his 1953 exchange with F.W. Bateson.


Leavis was a very great critic, whose essential subject, it seems to me, was language—its Kierkegaardian “appropriation,” its role in creating and maintaining values (both individual and communal), its maladies. He was way ahead of Heidegger and Gadamer.

But Winters was writing from the inside as a poet, a poet talking in part to other poets, including ones not yet in being, and Leavis was not.

And though Leavis could obviously have described formal features in formal terms (rather than reader-response ones) had he wished to, he chose not to, not least, I imagine, because it would have encouraged supposedly “objective” analyses, and the perception of texts as dualities in which “forms” existed independently of expressive utterance.

His influence on the writing of poetry, particularly given his preference for the major over the minor, and for a quasi-dramatic mode of utterance as a manifestation of an “inward” concern with life-values, whether in Shakespeare or Hopkins, or Donne, or Eliot, may not have been an unmixed blessing.

And it was, and maybe still is, insidiously easy to misapply, because misperceiving, what he was doing when he pounced on the “betraying” trope in “To Shakespeare.”


If that particular stretch of language caught the eye, it wasn’t in the sense that a dermatologist’s eye, scanning or skimming a smooth surface of healthy skin, is arrested by the blemish that signals an underlying, a perhaps incurable sickness—the way the body really is.

The weakness of the writing was there in the sonnet from the outset, though habituation to a particular style could have lulled you into missing it.

And such a lulling may be a necessary or at least a labour-saving mechanism when you have to read in bulk, and the writing, particularly if it’s free verse, isn’t flagrantly dreadful, and you’re waiting for something to catch your attention and slow you down.

But things go wrong when you assume that what you’re in search of is clues to the real, the innate “sensibility” of the writer, like a thought-police commissar waiting for betraying signs of someone’s true feelings about race, or gender, or social organization.

Or a religious heresy-hunter earlier.

The term “Leavisite,” with its implications of a puritanical narrowness, obviously didn’t fit writers for Scrutiny like D.W. Harding, Q.D. Leavis, L.C. Knights, W.H. Mellers, D.J. Enright, and others, any more than it now fits the wide-ranging mind of Michael Tanner—philosopher, music reviewer, Wagner freak.

But it did fit the odious H.A. Mason in his role as Scrutiny’s hatchet-man. (He literally almost never praised a writer, which makes you start imagining a song paralleling “That’s Entertainment,” the refrain of which would be, “That’s Standards!”) And it could be irritating or worse to come up against the kind of Downing-educated person who, while not running the risks attendant on celebrations (I mean, of works and authors outside the Scrutiny canon), assumed that he (it was normally a he) was sufficiently serving the life of the mind by—sniff! sniff!—detecting the ineradicable insufficiencies of sensibility in others.

Which no doubt contributed to the disastrous collapse of so-called practical criticism at Cambridge, meaning the making of intelligent comments on unfamiliar and unidentified poems or passages of prose, both fiction and non-fiction. And which left emerging graduates less equipped to deal with the deterioration of political discourse.


One does, of course, get impatient in one’s phrasing. This or that novelist, or poet, or movie-maker simply stinks, is third-rate, incorrigibly mediocre, fatally lacking in originality, and so on. Leavis himself spoke in that fashion at times. But it’s still shorthand.

And the superiority of Winters, there, is that he was always talking about texts, whether complete works or passages in works, and was judging those, and the quality of thinking and feeling in them, and not an abstraction called the “mind” of the writer.

The judging could go both ways, too. You might come upon something unexpectedly good in a work or oeuvre, as well as something more than usually bad.

It is an Aristotelian, not a Calvinist take on the world. What you do or say or write may be damnable at times. But it is damnable in itself, and not because it results from, and discloses, the fact that you are one of the damned.



Return to top