8

Associations

Another important way Graves influenced Empson was that he emphasized the special power that resides in the English language and that English poetry flaunts so spectacularly: the multiple, contradictory, supplementary, historical, and dynamically developing associations amongst the meanings of words. Empson concludes Seven Types of Ambiguity with a consideration of “the conditions under which ambiguity is proper,” and in doing so he foregrounds the word “associations”: “The methods I have been using seem to assume that all poetical language is debauched into associations to any required degree.”¹ He introduces the word via a passage by W.H. Auden and Cecil Day-Lewis in their preface to Oxford Poetry, 1927: “there is a ‘logical conflict, between the denotatory and connotatory sense of words; between, that is to say, an asceticism tending to kill language by stripping words of all associations and a hedonism tending to kill language by dissipating their sense under a multiplicity of associations.’”²

The word “associations,” like the word “conflict,” is especially associated in the 1920s with Graves, who fights all decade long to earn poetry’s “associative method of thought” a standing equal to that of the logic of “standard prose-writing,” which “only likes one meaning for every word … not three or four.”³ The word “association” is the one that Graves uses over and over again to explain the special nature of poetic language. In “The Illogic of Stoney Stratford and of Poetry” and “How Many Miles to Babylon?” (essays published in The Spectator in July of 1922), respectively, he defines poetry as “based on associative thinking” and asks readers to think about poems “on the lines of conflicting ideas reconciling themselves in symbolism by means of associative thought.”⁴ In his prefatory Note to On English Poetry, he describes “most scientists” as “insensitive to the emotional quality of words and their associative subtleties.”⁵ In his chapter on “Definitions,” he explains that “In poetry the implication is more important than the manifest statement; the underlying associations of every word are marshalled carefully.”⁶ He later describes the poet’s task as “the business of controlling the association-ghosts which haunt in their millions every word of the English language.”⁷ His chapter in Poetic Unreason on “The Illogical Element in Poetry” is a sustained criticism of the “scholastic system” that “makes no allowance for associative thinking, finds no virtue in a spoonerism or a pun suggested by the homophonic association of, say, ‘horse’ and ‘hoarse,’ or in fantastic slips either of pen or tongue.”⁸ He celebrates romantic poetry as “dependent on associative thought,” although he knows that from the point of view of the classical sensibility such poetry offers too many “illegitimate associations.”⁹ In a university textbook’s examples of breaches of logical rules, he finds “a clear example of associative thinking” and delights in identifying the “hidden associative tangle” caused by “the secondary meanings of words” and “the associations these meanings conjure up.”¹⁰ His point is that “associative thought is as modern and reputable a mode as intellectual thought.”¹¹

Empson understands this associative mode of thought to be enacted and invoked in the “linguistic form common in Shakespeare’s verse … ‘the (noun) and (noun) of (noun)’ … in which two, often apparently quite different, words are flung together, followed by a word which seems to be intended to qualify both of them.”¹² According to Empson, this linguistic form “is a powerful means of forcing [the reader] to adopt a poetical attitude toward words,” for “since this form demands that the reader should find a highest common factor of its first two nouns, it implies that he must open his mind to all their associations, so that the common factor may be as high as possible.”¹³ The often quite different words flung together by Shakespeare “have got to be associated by the constructive imagination of the reader.”¹⁴

Of course Coleridge – whose work Graves knew well and admired greatly – makes similar claims for the associative powers of the imagination, so Graves is by no means the first to champion associative thought. The contemporary figure who incites him to consider the virtues of associative thought, as Graves’s celebration of the meaningfulness of “fantastic slips either of pen or tongue” shows, is Freud.¹⁵ And as Empson points out, Freud’s ideas were available to young people interested in literary criticism from a number of sources: “Of course, you know everybody was talking about Freud and things like that. I might have got it from many other people who are now forgotten as well as Graves.”¹⁶ But certainly the most notorious theorist of associative thought in the 1920s was Graves, and everyone knew it. All of Graves’s books before his collaboration with Riding champion a psychoanalytical approach to the interpretation of literature, and even Richards pilloried him as representing the “gravest” dangers of Freudianism. “We all had it knocking about,” says Empson; but “the idea that you wanted to use Freudian theory on literary criticism somehow was the thing that Graves picked up a generation before me.”¹⁷

The tension between denotation and connotation to which Empson, Auden, and Day-Lewis refer is a version of the tension that Graves describes between the logic of standard prose writing and the associative mode of thought. He also describes it as the tension between logic and illogic, “analyses of cause and effect” and “associative thinking,” “intellectual thought” and the “emotional mode of thought,” classicism and romanticism.¹⁸ The ideal for the one sensibility is poetry that is a “logically harmonious and rational statement,” embodying thinking as “a logical process of one continuous strand”: “the poetry recommended was to pursue a steady course in which every image was to be deduced from the previous one, the whole scheme was to work out according to the logic of human nature and other universal characteristics, and there was no place to be left for caprice.”¹⁹ The ideal here is “direct grammatic statement” and “self-aware allegory”; ultimately, “one meaning for every word.”²⁰ The ideal for the other sensibility is “secondary meanings” for every word, poetry whose grammar is “of sibylline obscurity,” and allegory that is “latent.”²¹ The “secondary meanings of words, however remote, react on each other in moments of passion if there is any similarity of emotional disturbance between the associations these meanings conjure up”; “the grammatic construction is continually being interrupted by fresh images before the others have submitted themselves for logical classification”; poetry of this sort, “appearing in a capricious unforeseeable way,” is “bound up with allegories interlacing by homophonic and other [logically] illegitimate associations.”²²

So must it be in poetry that is not “a condensation and rearrangement of past events, according to a preconceived logical structure,” but rather “a new entity”: new because it expresses a conflict that is in the process of being experienced and expresses opposed ideas that are in the process of being linked. And this linkage takes place not by means of a timeless relationship supported by the logic of cause and effect but according to a temporal relationship emerging from the poet’s experience and supported by the illogic of homophonic caprice and tongue-slipping accident, or any number of other aspects of the two-faced machinery of poetic language.²³ Such poetry evidences “inconsistency of plot” ( “things happen which realistically-minded strangers find difficult to understand”), “geographical confusion,” “inconsistent character” ( “characters are continually changing from what they at first represented into something else”), and “the development of this emblem or that in unforeseen directions.”²⁴

Auden and Day-Lewis are concerned to find a way between “an asceticism tending to kill language by stripping words of all association and a hedonism tending to kill language by dissipating their sense under a multiplicity of associations.”²⁵ Similarly, Empson is concerned to balance, on the one hand, the impression that by his method “all poetical language is debauched into associations to any degree required” with, on the other hand, the impression that he offers a “decent homage to the opposing power.”²⁶ All three reflect Graves’s concern in Poetic Unreason to find the middle way between the same extremes: “if it be … claimed that the logical method is the only right method, and all others useless because illogical, I would point out that the romantic method has a similar scorn for the logical method: both may therefore be suspect when they advance these mutually exclusive claims.”²⁷

According to Graves, every poet – whether of the classical logical tradition or the romantic associative tradition – depends on both modes of thought. Although the devotee of logic expects statements to be “empirically confirmable by certain sensuous tests based on a view of the absolute reality of the spatio-temporal structure,” and the associationalist nurtures the “madness and contradiction” of “poetic metamorphosis” by which one thing becomes another “so that the reality of the spatio-temporal structure being challenged ceases to be absolute reality,” their scorn for each other belies what is common in their world of experience:

the Spatio-temporally-minded logician will theorize about ghosts and fairies as subjective phenomena while denying their objective existence, and the romantic or mystic will employ spatio-temporal formulae in an account of his visions while denying the adequacy of the senses to record the experience. And in the same way as a poet or transcendentalist will be dependent on the railway time-table as the most cramped logician, so this logician will night after night permit himself a series of the most fantastic dreams, and unwittingly derive as much benefit from them in aid of his spatio-temporal logic as the time-table gives the transcendental in regulating his journey to some grand conference of international mystics.²⁸

And so despite the apparent rivalry between classical and romantic, logical and illogical, denotative and connotative sensibilities, the history of English poetry has always involved the expression of both sensibilities:

though they are both expressions of a mental conflict, in Classical poetry this conflict is expressed within the confines of waking probability and logic … in Romantic Poetry the conflict is expressed in the illogical but vivid method of dream-changings … The dream origin of Romantic Poetry gives it … a naturally hypnotic effect. [Classical] poetry … feels the need of this easy suggestion to the audience for ideal thinking; and finds it necessary to avoid realism by borrowing shreds of accredited metamorphic diction and legend and building with them an illusion of real metamorphism … The borrowed Metamorphism is hardened to a convention and a traditional form … Sometimes, however … a nation’s Classical tradition is broken by popular ridicule and the reappearance of young Metamorphic Poets. But after a little paper-bloodshed … the Classical tradition reappears, dressed up in the cast-off finery of the pioneer Metamorphics … This complicated dog-eat-dog process is cheerfully called “The Tradition of English Poetry.”²⁹

The way forward is to recognize that these two modes of thought are complementary strategies for dealing with conflict:

once it is admitted by philosophers that though Romantic thought cannot be exactly foreseen, neither any more can intellectual thought; and further, when they admit that emotional poetry and thought, once it has appeared, has discernible an orderly orientation and inevitability; and when further they observe that such poetry and thought is not so primitive in the popularly accepted sense that it cannot use intellectual formulae for its own purposes, but is a method constantly employed unknown to themselves by the most sophisticated and Aristotelian minds, and of great service to them – then, finally, the clash between Classical and Romantic can end and these terms can be used to qualify the forms of poetry alternately appearing as either one or other method of thought employed for dealing with any conflict.³⁰

Conflict is a human condition, and so is the fact that our minds proceed by both the emotional and the intellectual modes of thought. Just as poetic language cannot confine itself to either of these extremes, neither can the common language of the person in the street.

Empson agrees. He recognizes contemporary fears “about a breakup of the English language now visible on the hoardings, where words abandon grammatical distinction and work by association,”³¹ but is not worried: “This may be alarming but it is not a novelty; it may restore to the language a flexibility it has not possessed since the days of its greatest triumphs.”³² The Elizabethans enjoyed a golden age of English, and he offers the relationship between Shakespeare and his Elizabethan readers as a model of the proper balance between the associative and the intellectual modes of thought in the English language. Shakespeare, he says, will allow him “To show how extremely flexible it [language] was then; how much, so long as it can be combined both with the possibility and encouragement of intellectual precision, an extreme flexibility of language is to be valued.”³³

•

Graves emphasizes the special powers of the English language for comprehending and expressing this double aspect of the English sensibility even before he publishes his first book of literary theory. He concludes On English Poetry by reproducing his 1921 letter to the editor of the Tracts of the Society for Pure English in which he argues that a lack of associations among words is as dangerous to accurate expression as an excess: “I would like to sound a warning against the attempt to purify the language too much – ‘one word, one meaning’ is as impossible to impose on English as ‘one letter, one sound.’”³⁴ Of course “one word, one meaning” is the extreme that Graves associates with scientists, philosophers, and logicians, and his books and essays will soon campaign vigorously against the hegemony of these linear thinkers. Nonetheless, Graves agrees that “a common-sense precision in writing is clearly necessary.”³⁵ Without a certain degree of precision, something so simple as contemplating an orange can lead to a chaos of competing sounds and meanings: “Old gentlemen usually pare their oranges, but the homophonic barrage of puns when Jones père prepares to pare a pair of – even oranges (let alone another English-grown fruit), has taught the younger generations either to peel a norange or skin their roranges.”³⁶ Still, Graves insists that “over-definition” of a word “discourages any progressive understanding of the idea for which it acts as a hieroglyph.”³⁷

This “progressive understanding” of an idea comes from the word’s associations. The orange can be anatomized in terms of “seeds,” “calix,” “exocarp, carpel, and ovule,” and it can be eaten by “paring the integument and afterwards removing the divisions of the fruit for mastication.”³⁸ Yet there is an other than “literary and semi-scientific language” for oranges according to which seeds are pips; the withered calyx is kim; the white part under the rind is the blanket or kill-baby. Anatomizing and eating an orange according to this more homely language, one must peel the “skin off the rind, ignoring the kim and scraping away the kill-baby, then pull out the pigs, chew them decently, and put the pips to their proper use.”³⁹ Reducing the definition of “orange” to scientific denotation leaves out of account the way the word is now embedded in non-scientific networks of meaning and potentially prevents its progressive association with other networks of meaning not yet anticipated. And so, “the more precisely circumscribed a word, the less accurate it is in relation to other closely-defined words” – or, in other words, “good English surely is clear, easy, unambiguous, rich, well-sounding, but not self-conscious; for too much pruning kills.”⁴⁰

Empson develops very similar points about the vibrancy of contemporary English in the last chapter of Seven Types of Ambiguity. Perhaps remembering the particular complaint that Graves makes about the scientist who is so inaccurate in his use of language as to give a grub a hand by describing it as manufacturing a channel (the scientist, says Graves, uses words as “lifeless counters, weights, measures, or automatic engines wrongly adjusted”), but certainly remembering at least the general kind of debate that the Society for Pure English was stirring, Empson observes that “people sometimes say that words are now used as flat counters, in a way which ignores their delicacy; that English is coming to use fewer of its words, and those more crudely.”⁴¹ Oranges and grubs alike suffer from the poor language skills of Graves’s scientist: oranges should be described by more words; grubs should be described by more accurate words.

Empson perhaps recalls Graves’s picture of the “good English” anatomy of an orange being displaced by “semi-scientific language” in his own comments on science’s general impact on language: “The sciences might be expected to diminish the ambiguity of the language, both because of their tradition of clarity and because much of their jargon has, if not only one meaning, at any rate only one setting and one point of view. But such words are not in general use; they act only as a further disturbing influence on the words used already.”⁴² Graves’s phrase for the scientist’s ambition for language – “one word, one meaning” – echoes here in Empson’s emphasis on the “one meaning,” “one setting,” and “one point of view” that science prefers. Similarly, Empson’s suggestion that the singular definition of a scientific term nonetheless becomes part of the network of words in a language so as potentially to “act … as a further disturbing influence on the words used already” echoes Graves’s point that the “secondary meanings of words, however remote, react on each other in moments of passion if there is any similarity of emotional disturbance between the associations these meanings conjure up.”⁴³

Empson grants that what he calls ambiguity (and what Graves calls secondary meaning) is not just a property of poetic language. He points to the typical newspaper headline ( “given particular meaning” by ambiguous words that “can be either nouns or verbs, and would take kindly to being adjectives”) and general “journalistic flatness” (in which “the word is used … to stand for a vague and complicated mass of ideas and systems which the journalist has no time to apprehend”) as the context for his suggestion that “it would be easy enough to take up an alarmist attitude, and say that the English language needs nursing by the analyst very badly indeed.”⁴⁴ Yet like Graves, who complains that “It is no longer practical to coin words, resurrect obsolete ones and generally to tease the language as the Elizabethans did,” Empson argues that the headline can be “a very effective piece of writing” because

it conveys its point … with a compactness which gives the mind several notions at one glance of the eye, with a unity like that of metaphor … Nor can I feel that it will be a disaster if other forms of English literature adopt this fundamental mode of statement, so interesting to the logician; it is possible that a clear analysis of the possible modes of statement, and a fluid use of grammar which sets out to combine them as sharply as possible into the effect intended, may give back something of the Elizabethan energy to what is at present a rather exhausted language. But this is by the way; I have only used the headline to insist on the value of a detached and willing attitude to grammar, to show that the grammatical sentence is not the only form of statement in modern English.⁴⁵

Whether in the secondary meanings that attend talk about oranges or in the syntactical and semantic ambiguities of newspaper headlines, Graves and Empson find a potential pattern of associations awaiting only the wit of a modern Elizabethan to spy out and express.

•

The key idea that Graves passes along to Empson concerns the pattern of associations that secondary meanings constitute. This idea is the basis of Empson’s distinction between “accidental ambiguity” ( “English is becoming an aggregate of vocabularies only loosely in connection with one another, which yet have many words in common, so that there is much danger of accidental ambiguity, and you have to bear firmly in mind the small clique for whom the author is writing”) and “two-faced implications of any plausibility.”⁴⁶ On the one hand, “An ambiguity is not satisfying in itself, nor is it, considered as a device on its own, a thing to be attempted; it must in each case arise from, and be justified by, the peculiar requirement of the situation.”⁴⁷ On the other hand, each ambiguity must support a larger whole: “all the subsidiary meanings must be relevant, because anything (phrase, sentence, or poem) meant to be considered as a unit must be unitary, must stand for a single order of mind.”⁴⁸ As Graves writes in explanation of the emotional conflict at the heart of poetry, “I plead the rule that ‘Poetry contains nothing haphazard,’ which follows naturally on the theory connecting poetry with dreams. By this rule I mean that if a poem, poem-sequence or drama is an allegory of genuine emotional experience and not a mere cold-blooded exercise, no striking detail and no juxtaposition of apparently irrelevant themes which it contains can be denied at any rate a personal significance – a cipher that can usually be decoded from another context.”⁴⁹ There is always at least the unity of a psycho-biographical associational logic recoverable from every proper poem. But the fact that an element in a poem has more than one meaning is interesting only if there is a whole within which the various meanings cohere.

As early as his 1921 letter to the editor of the Society for Pure English, Graves declares his belief that language is a living entity and implicitly an organic whole. Reviewing proposals made in the Society’s first five tracts, he figures the English language as a thriving eco-system (comprising weeds, flowers, birds, and bees) and warns against a pedantic, mechanical, technological attitude toward language that could prove lethal: “By all means weed out homophones, and wherever a word is overloaded and driven to death let another bear part of the burden; suppress the bastard and ugly words of journalese or commerce; keep a watchful eye on the scientist, take necessary French and Italian words out of their italics to give them an English spelling and accentuation; call a bird or flower by its proper name, revive useful dialect or obsolescent words, and so on; that is the right sort of purification, but let it be tactfully done, let the Dictionary be a hive of living things and not a museum of minutely ticketed fossils.”⁵⁰ Since “too much pruning kills,” the pruners ought at least to be the ones with green thumbs and not those “using words … as lifeless counters, weights, measures, or automatic engines wrongly adjusted.”⁵¹ Auden and Day-Lewis follow the same argument: too much pruning in favour of denotation kills; a choking growth of connotative weeds also kills.

Graves is certainly interested in language understood diachronically – he is curious about the derivation of words and their changing uses over time (of the word “pig” as used for the divisions of an orange, he wonders, “is the derivation from the tithe- or parson’s pig known by its extreme smallness?”) but he is more interested in the way words are implicated synchronically in the living system of the English language: “As one rather more interested in the choice, use, and blending of words than in the niceties of historical grammar, and having no greater knowledge of etymology than will occasionally allow me to question vulgar derivations of place-names, I would like to sound a warning against the attempt to purify the language too much – ‘one word, one meaning’ is as impossible to impose on English as ‘one letter, one sound.’”⁵²

According to Graves, any verbal expression is a temporal and temporary event. It harnesses the various meanings available in the service of the expression of an idea, but such an idea is no more fixed and final than the meanings of each of the individual words or the particular combination of words by which it has come to be expressed. Understanding is always in progress, and its progress is enabled by the connections between ideas revealed by the connections between the words that express these ideas. And so: “Over-definition … discourages any progressive understanding of the idea for which it acts as hieroglyph. It even seems that the more precisely circumscribed a word, the less accurate it is in relation to other closely-defined words.”⁵³

As we have seen, the Society for Pure English letter culminates in a demonstration of the “progressive understanding of the idea” represented by the word “orange.” Graves begins by telling the “story of a governess” who is “suspicious” of her female charges’ “scientific” elder brother’s description of the earth as “an oblate spheroid” and suggests to them that it “is nicer for little girls to say that the earth is more or less the shape of an orange.” His paragraph’s tour de force deserves quoting at length:

From [this] fruit … can be drawn our homely moral of common sense in the use of words. As every schoolboy I hope doesn’t know, the orange is the globose fruit of that rutaceous tree the citrus aurantium, but as every schoolboy certainly is aware, there are several kinds of orange on the market, to wit the ordinary everyday sweet orange from Jaffa or Jamaica, the bitter marmalade orange that either comes or does not come from Seville, the navel orange, and the excellent “blood,” with several other varieties. Moreover the orange has as many points as a horse, and parts or processes connected with its dissection and use as a motorcycle. “I would I were an Orange Tree, that Busie Plante,” sighed George Herbert once. I wonder how Herbert would have anatomized his Orange, then a rarer fruit than today when popular affection and necessary daily intercourse have wrapped the orange with a whole glossary of words as well as with tissue-paper … Peel (subst.) is ousting rind; a pity because there is also peal as a homophone; but I am glad to say that what used to be called divisions are now almost universally known as fingers or pigs … the seeds are “pips,” and quite rightly too, because in this country they are seldom used for planting, and “pip” obviously means that when you squeeze them between forefinger and thumb they are a useful form of minor artillery; then there is the white pithy part under the outer rind; I have heard this called blanket, and it is pretty good, but I have also heard it called kill-baby, and that is better; for me it will always remain kill-baby. On consulting Webster’s International Dictionary I find that there is no authority or precedent for calling the withered calix on the orange the kim, but I have done so ever since I can remember, and have heard the word in many respectable nurseries (it has a fascination for children), and I can’t imagine it having any other name. Poetical wit might call it “the beauty-patch on that fairy orange cheek”; heraldry might blazon it, on tenne, as a mullet, vert, for difference; contemporary slang would probably explain it as that “rotten little star-shaped gadget at the place where you shove in your lump of sugar”; but kim is obviously the word that is wanted, it needs no confirmation by a Dictionary Revisal Committee or National Academy. There it is, you can hardly get away from it.⁵⁴

According to Graves, the idea of just what an orange has been, is, and shall be makes itself available continually and continuously to the progressive understanding of speakers of English – from schoolboys to correspondents of the Society for Pure English – by the fact that it is circumscribed (not overly precisely, but precisely enough) by so many ordinary words with so many associations and secondary meanings.

In conclusion, Graves epitomizes the “clear, easy, unambiguous, rich, well-sounding” way that good English works by proper instruction on how to eat such an orange. Such instruction must find the mean between the extremes of the “pedantic scientist,” on the one hand, and “contemporary slang,” on the other. It certainly ought not to be offered in the form of a “self-conscious” and “semi-scientific language” “insisting on paring the integument and afterwards removing the divisions … for mastication.” Rather, the instruction should be in the form of a language as common to the boys and girls of the “respectable nurseries” as to “the man in the street.”⁵⁵

•

In On English Poetry, written at about the same time as this letter but with a publication date a year later (1922), Graves explains the application of this perspective on “good English” to poetry. In a prefatory comment to his letter to the Society for Pure English, he indicates that it “explains [his] attitude to the careful use of English by prose writers as well as by poets,” and he suggests that the letter be “read in conjunction with” his chapter on poetic diction: “the poet will be best advised to choose as the main basis of his diction the ordinary spoken language of his day; the reason being that words grow richer by daily use and take on subtle associations which the artificially bred words of literary or technical application cannot acquire with such readiness.”⁵⁶

Elsewhere in On English Poetry, as we have seen, he declares that these subtle associations form patterns in good poetry:

The underlying associations of each word in a poem form close combinations of emotion unexpressed by the bare verbal pattern … In this way the poet may be compared with a father piecing together a picture-book puzzle for his children. He surprises them at last by turning over the completed picture, and showing them that by the act of assembling the scattered parts of “Red Riding Hood with the Basket of Food” he has all the while been building up unnoticed underneath another scene of tragedy – “The Wolf eating the Grandmother” … The analogy can be more closely pressed; careless arrangement of the less important pieces or wilfully decorative borrowing from another picture altogether may look very well in the upper scene, but what confusion below! … The possibilities of this pattern underneath have been recognized and exploited for centuries in Far Eastern systems of poetry. I once even heard an English Orientalist declare that Chinese was the only language in which true poetry could be written, because of the undercurrents of allusion contained in every word of the Chinese language. It never occurred to him that the same thing might be unrecognizedly true also of English words.⁵⁷

Graves’s early criticism is devoted to making this recognizedly true.

His clearest formulation of the special poetic nature of the English language appears in the book that Empson later said had most influenced him: Impenetrability: Or The Proper Habit of English (1926). Graves regards English as a language that employs “the imaginative method,” unlike the “more logical languages” such as French, which is now “steady enough and secure enough against imaginative interruption.”⁵⁸ He shows what he means by considering the adjective “brazen”: “All the qualities … which the imagination can give to brass are latent in the word.”⁵⁹ “‘Brazen’ may mean ‘shining with a metallic glare’ … Or it may mean ‘strident,’ by recollection of the noise that a brazen trumpet or bugle makes … Or it may mean ‘cased with brass,’ and therefore ‘shameless and insolent’ … It may mean ‘of less worth than gold.’”⁶⁰ According to Graves, “It is the persistent use of this method of ‘thought by association of images’ as opposed to ‘thought by generalised preconceptions’ that distinguishes English proper from the more logical languages.”⁶¹ The method of English as a language, therefore, is fundamentally cognate with the special nature of poetic language.

In Impenetrability, Graves makes explicit his belief that language – especially English – is capable of signifying patterns of meaning beyond the author’s conscious awareness and deliberate control. Although “for French prose-writers and many celebrated French verse-writers the conceits of the words they use are dulled and show no rebellious tendency to form illicit assemblies that might affect the argument,” the situation in English literature is quite the opposite: “For English writers of prose or verse, so soon as a gust of natural feeling snatches away the typographical disguise in which their words are dressed, the conceits appear in all freedom: at first they enliven and enforce the argument, but after a while, if the author is not wary, they desert it and begin a digressive dance of their own.”⁶²

In the passage of Graves’s close literary analysis that Empson pointed out to Riding more than forty years after he first read it, Graves articulates the attitude toward language and literature that Empson explains at greater length in Seven Types of Ambiguity. Graves begins by comparing English to Latin and its derivative vernaculars such as Italian and French, observing that “When particular words very highly charged with meaning in their context occur in English literature, this is counted a great virtue. In logical literatures it is a vice, by the rule of ‘one word, one meaning.’”⁶³ Empson was clearly paying attention, eventually saying virtually the same thing: “a poetical use of ambiguity … has some claim to be considered native to the language. I really do not know what importance it has in other European languages … my impression is that while it is frequent in French and Italian, the subsidiary meanings are nearly always bad grammar, so that the inhabitants of those countries would have too much conscience to attend to them.”⁶⁴ Graves then proceeds to explicate the highly charged meanings of the word “clasped” in Keats’s “The Eve of St Agnes,” the example that Empson later recalled for Riding, and concludes: “English is strong enough to bear any weight so long as the load is properly packed and adjusted. Its greatest virtue as a literary language is its readiness to absorb foreign and technical words, and words from the better sort of slang, and to extemporise not only new imaginative phrases but distinguishable varieties of any overworked or ambiguous word.”⁶⁵

Having demonstrated how the effect of certain lines of poetry depends upon words with two or three meanings, Graves declares that “This … is the constant practice of those English poets who achieve the most admired phrases. The reader is not rationally aware of the principle underlying such phrases: he knows that they delight him but does not in the act of reading poetry dissect them.”⁶⁶ Empson’s point is the same: “I have been trying to analyse verses which a great variety of critics have enjoyed but only described in terms of their effects; thus I have claimed to show how a properly-qualified mind works when it reads verses, how those properly-qualified minds have worked which have not at all understood their own working.”⁶⁷

•

As always, one can trace Graves’s influence on Empson in ways both big and small. For example, Graves observes that “English proper has always been very much a language of ‘conceits’; that is, except for the purely grammatic parts of speech, which are in general colourless enough, the vocabulary is not fully dissociated from the imagery from which it developed: words still tend to be pictorial and not typographic … For English writers of prose or verse, so soon as a gust of natural feeling snatches away the typographical disguise in which their words are dressed, the conceits appear in all freedom: at first they enliven and enforce the argument, but after a while, if the author is not wary, they desert it and begin a digressive dance of their own.”⁶⁸ Empson identifies the same special claim for English as a poetic language, he notes the same danger that the unwary face in the handling of it, and he even avails himself of the same metaphor: “All languages are composed of dead metaphors as the soil of corpses, but English is perhaps uniquely full of metaphors of this sort, which are not dead but sleeping, and, while making a direct statement, colour it with an implied comparison. The school rule against mixed metaphor … is largely necessary because of the presence of these sleepers, who must be treated with respect; they are harder to use than either plain word or metaphor because if you mix them up you must show you are conscious of their meaning, and are not merely being insensitive to the possibilities of language.”⁶⁹

Similarly, at three different points in On English Poetry, Graves characterizes conflict within the poet as analogous to conflicts like war and civil disturbance: “The mind of a poet is like an international conference composed of delegates of both sexes and every shade of political thought, which is trying to decide on a series of problems of which the chairman has himself little previous knowledge”; “I have … attempted to show Poetry as the Recorder’s précis of a warm debate between the members of the poet’s mental Senate on some unusually controversial subject”; “new ideas troop quietly into [the poet’s] mind until suddenly every now and again two of them violently quarrel and drag into the fight a group of other ideas that have been loitering about at the back of his mind for years; there is great excitement, noise and bloodshed, with finally a reconciliation and drinks all round. The poet writes a tactful police report on the affair and there is the poem.”⁷⁰ And here again it is clear that Empson paid close attention, writing in his turn: “In Shakespeare’s great parades of associations the attendants are continually quarrelling among themselves”; and “The Folio’s comma … heightens the civil war in the line by dividing it in two.”⁷¹

1 William Empson, Seven Types of Ambiguity: A Study of Its Effects in English Verse, 1st ed. (London: Chatto and Windus, 1930; New York: Harcourt, Brace and Company, 1931), 296.

2 W.H. Auden and Cecil Day-Lewis, preface to Oxford Poetry (1927), quoted by Empson, Seven Types of Ambiguity, 1st ed., 296. Blending ideas from Eliot’s “Tradition and the Individual Talent” and “The Metaphysical Poets” with Graves’s “conflict theory,” Auden and Day-Lewis also foreground the idea of poetry as an experience of psychological “conflict” that produces a “new harmony”: “Emotion is no longer necessarily to be analysed by ‘recollection in tranquillity’; it is to be prehended emotionally and intellectually at once. And this is of most importance to the poet; for it is his mind that must bear the brunt of the conflict and may be the first to realize the new harmony which would imply the success of this synchronization.” See the preface to Oxford Poetry (1927), in The Complete Works of W.H. Auden: Prose and Travel Books in Prose and Verse: Volume 1. 1926–1938, ed. Edward Mendelson (Princeton: Princeton University Press, 1997), 3.

3 Robert Graves, Poetic Unreason: And Other Studies (London: Cecil Palmer, 1925), 119; On English Poetry: Being an Irregular Approach to the Psychology of This Art, from Evidence Mainly Subjective (London: William Heinemann, 1922), 14; and Poetic Unreason, 120.

4 Robert Graves, “The Illogic of Stoney Stratford and of Poetry,” Spectator, 129 (15 July 1922): 87; Graves, “How Many Miles to Babylon? An Analysis,” The Spectator, 129 (22 July 1922): 117.

5 Graves, On English Poetry, viii.

6 Ibid., 14.

7 Ibid., 71.

8 Graves, Poetic Unreason, 117.

9 Ibid., 118, 132.

10 Ibid., 119, 120, 120.

11 Ibid., 127.

12 Empson, Seven Types of Ambiguity, 1st ed., 114.

13 Ibid., 115.

14 Ibid., 116.

15 Graves, Poetic Unreason, 117.

16 Empson, quoted by Haffenden, Selected Letters of William Empson, ed. John Haffenden (Oxford: Oxford University Press, 2006), 431n10.

17 Ibid.

18 Graves, Poetic Unreason, 117, 117, 127, 127.

19 Ibid., 121, 131, 132.

20 Ibid., 125, 125, 120.

21 Ibid., 120, 12, 125.

22 Ibid., 120, 135, 132.

23 Ibid., 124

24 Ibid., 123, 125, 124, 124, 125, 123–4.

25 Auden and Day-Lewis, preface to Oxford Poetry, 3.

26 Empson, Seven Types of Ambiguity, 1st ed., 296.

27 Graves, Poetic Unreason, 118.

28 Ibid., 137, 137, 135, 137, 137–8.

29 Graves, On English Poetry, 73–5.

30 Graves, Poetic Unreason, 133.

31 Empson, Seven Types of Ambiguity, 1st ed., 111.

32 Ibid.

33 Ibid.

34 Graves, On English Poetry, 143.

35 Ibid., 146.

36 Ibid., 147.

37 Ibid., 146.

38 Ibid., 148.

39 Ibid., 148–9.

40 Ibid., 146, 149.

41 Ibid., 144; Empson, Seven Types of Ambiguity, 1st ed., 299.

42 Empson, Seven Types of Ambiguity, 1st ed., 299.

43 Graves, Poetic Unreason, 120.

44 Empson, Seven Types of Ambiguity, 1st ed., 299, 300, 299, 299, 298.

45 Ibid., 300–1.

46 Ibid., 299, 299, 306.

47 Ibid., 297–8.

48 Ibid., 296.

49 Graves, On English Poetry, 32–3.

50 Ibid., 143–4.

51 Ibid., 149, 145.

52 Ibid., 147, 143.

53 Ibid., 146.

54 Ibid., 147–8.

55 Ibid., 148–9.

56 Ibid., 143, 42.

57 Ibid., 24–5.

58 Robert Graves, Impenetrability; or the Proper Habit of English (London: The Hogarth Press, 1926), 40.

59 Ibid., 39.

60 Ibid., 39–40.

61 Ibid., 40.

62 Ibid., 11.

63 Ibid., 55.

64 Empson, Seven Types of Ambiguity, 1st ed., 86–7.

65 Graves, Impenetrability, 55–9.

66 Ibid., 57.

67 Empson, Seven Types of Ambiguity, 1st ed., 314.

68 Graves, Impenetrability, 9–11

69 Empson, Seven Types of Ambiguity, 1st ed., 32–3.

70 Graves, On English Poetry, 33–4, 113, 26.

71 Empson, Seven Types of Ambiguity, 1st ed., 118–19, 119.