1. Suggestion and Statement in Poetry Krishna Rayan (english Poetry)

Suggestion

and

Statement

in

Poetry

KRISHNA RAYAN

UNIVERSITY OF LONDON

THE ATHLONE PRESS

1972

Published

by

THE

ATHLONE

PRESS

UNIVERSITY

OF

LONDON

at

4

Gower

Street,

London

WC1

Distributed

by

Tiptree

Book

Services

Ltd

Tiptree,

Essex

©

Krishna

Rayan

1972

0

485

11134

9

Printed

in

Great

Britain

by

T.

&

A.

CONSTABLE

LTD

EDINBURGH

PREFACE

In this study of suggestion, in relation to statement, in

English poetry, I have started by examining 'Suggestion —

the term and the concept: sketching its origin and develop-

ment and indicating its present status. I have proceeded, in

the third, fourth and fifth chapters, to consider the two kinds

into which suggestion can be divided: the suggestion of

emotion, and suggestion through metaphor. The sixth and

seventh chapters, to be read with the fourth, analyse sug-

gestion and statement in relation to each other. Two case

studies follow: T. S. Eliot and Wordsworth. We next revert

to character and plot as objective correlatives. The con-

cluding chapter touches upon certain problems of suggested

meaning, such as: Should it be evaluated? When is it para-

phrasable? What happens when it is defined or explicated?

I have constantly brought in concepts from Sanskrit

poetics, but invariably as points of departure for discussion

of poetic practice and critical ideas in English. If at certain

points the book sounds like special pleading for Sanskrit

theory, I can only say that that, at any rate, was not my

intention. This essay is not an exposition of Sanskrit poetics,

nor is it a comparative study of Sanskrit and English critical

theory—anything in that line will have to await the advent

of the super-scholar who matches a knowledge of what the

ninth-to-eleventh-century Indian theorists of poetry meant

with a specifically modern sensibility and equipment which

can enable him to work at a deeper level than that of seem-

ing affinities and to speak a language which will not distort

vi

PREFACE

what those masters said and yet will be comprehensible to

present-day English-speaking readers.

In citing the work of other critics I suspect that I may be

charged with having drawn upon too small a company and

then on too few of their writings—Abercrombie, Tillyard,

Richards, Empson, Eliot, Cleanth Brooks, Wimsatt, and

John Bayley. Here is a selection at least part of which will

be felt to be either old-fashioned or eccentric. But if certain

critical works seemed to offer more ideas than others,

serving as useful points of departure, I saw no reason why

I should not make use of them freely.

There are other and more considerable debts owed to

those whose support and advice made this book possible:

F. W. Bateson, Bruce Pattison, Harold Osborne, Robin

Skelton and Krishna Arjunwadkar.

K. R.

ACKNOWLEDGEMENTS

I am grateful to the editors and publishers of the following journals for permission to reprint material originally published in them (date of numbers is shown in brackets):

Essays in Criticism for Chapter 8 (July 1969) and part of Chapter 9 (October 1967); British Journal of Aesthetics for part of Chapter 2 (January 1969) and Chapter 3 (July 1965);

Malahat Review for Chapter 5 (April 1967), Chapter 6 (April 1970) and part of Chapter 11 (July 1968).

Acknowledgement is also made to the following for use of copyright material as specified in detail in the relevant notes: Faber and Faber Ltd for extracts from poems by Thom Gunn and Robert Lowell; Granada Publishing Ltd for extracts from poems by R. S. Thomas; the Longman Group Ltd for a poem by Patricia Beer; Macmillan and Co. Ltd for a poem by W. B. Yeats; Routledge and Kegan Paul Ltd for a poem by John Holloway.

K. R.

CONTENTS

1 Suggestion Today

1

2 Suggestion: From Poe to the Present

17

3 Suggestion through the Objective Correlative

31

4 The Lamp and the Jar: Stated and Suggested Meaning

52

5 Suggestion through Metaphor

67

6 Stating and Suggesting by Turns

86

7 Statement Poetry

97

8 Suggestion as a Classical Method

121

9 Suggestion or Statement? The Case of Wordsworth

134

10 Suggestors of Emotion

144

11 Notes on Suggested Meaning

154

Notes

172

Index

179

1

Suggestion Today

Finding that the fallen trees obstruct the view, you leave the

main path and advance to the open cliff edge where a close-

up of the Victoria Falls must be easy to have. As you

approach the coign, however, you see a giant cloud of spray

climbing out of the gorge, and the next moment you are

blinking at a grey impenetrable wall that has obliterated

everything. Nevertheless, the Falls are a commanding

presence, for through the barrier of spray you can hear a

muffled roar, and you know that the massed waters are

tumbling into the gorge at your feet. The evidence of your

ears affirms that the Victoria Falls are there before you, but

their identity is indeterminate, ambiguous and obliquely ap-

prehended—you are aware that behind the spray it can

either be an angry torrent gushing through a constriction,

or a stately wall of water; it can either be a row of shrunken

streams, or a single undiminished expanse. The first man to

be able to define this apprehension of the Zambesi’s veiled

descent was a local Lozi whose nomenclatory imagina-

tion called the Falls Musi-o-tunya, ‘the Smoke that

Thunders’.

Here in this sightless instant—you tell yourself as you

retreat before the drenching spray back to the main path—is

held an experience that uniquely projects the whole

process of poetic suggestion. Like the Falls, the suggested

meaning waits beyond the frontiers of the overt and the

I

2

SUGGESTION TODAY

explicit, unseen but powerfully felt; and its nature is im-

precise, manifold, indirectly known. We wouldn’t know it

but for the element of statement present—in this case, the

rumble of the Falls. But the statement is necessarily frag-

mentary; the principle of omission, which is what the spray

served when it cancelled the visual component of the

presentation, is central. Without suppression there can

be no suggestion—you realize this almost at once, for

the spray has subsided as abruptly as it had risen, and

through a wide parting in the trees the Victoria Falls

reveal their entire architecture of rock and torrent,

every detail of shape, colour and movement lucid in the

late August sun. This is impressive too, but in a different

way.

Not only the phenomenon of the Victoria Falls with their

curtain of spray, but the very name—Musi-o-tunya—re-

produces the anatomy of suggestion. Metaphor is one of the

prime movers of suggestion; the two metaphors for the

spray and the sound, making up the whole name, eloquently

render this truth. And the coupling of thunder and smoke

(strangely reminiscent, across cultures and maybe centuries,

of Mallarmé’s tonnerre et rubis) exemplifies, generally, the

output of unspoken meaning when discordant images are

forcibly drawn together, and specifically, the action of that

powerful lever of suggestion, synaesthesia.

The Victoria Falls are as good a case as any of nature

imitating art. But poetic suggestion is indeed so many things

at once or by turns that not even the Victoria Falls, for all

their variety and complexity, will do as a model. Even to

form a first rough idea of suggestion, we have to get away

from the waterfall metaphor for poetry to poetry itself. To

start with the aspect we have already looked at, suggestion

SUGGESTION TODAY

3

can be, for one thing, suppression, resulting in truncated representation or fragmentary syntax:

But once upon a time

the oakleaves and the wild boars

Antonio Antonio

the old wound is bleeding.

(Herbert Read, 'Cranach')

Suggestion can be—to mention another of its many shapes and the one that is most suspect today—incantation, the eloquence of sounds and rhythms:

For the stars and the winds are unto her

As raiment, as songs of the harp-player;

For the risen stars and the fallen cling to her,

And the southwest-wind and the west-wind sing.

(Swinburne, Atalanta in Calydon)

Or instead of the sound values of words, their opulent semantic associations can be exploited with a view to suggestiveness:

A grief ago,

She who was who I hold, the fats and flower,

Or, water-lammed, from the scythe-sided thorn,

Hell wind and sea,

A stem cementing, wrestled up the tower,

Rose maid and male,

Or, masted venus, through the paddler's bowl

Sailed up the sun.

(Dylan Thomas, 'A grief ago')

Suggestion can also work by realizing the alternative meanings, as distinct from the multiple associations, of words,

4

SUGGESTION TODAY

such as 'mede', 'peynted' and 'proces' in the following stanza from Troilus and Criseyde:

What? is this al the Ioye and al the feste?

Is this your reed, is this my blisful cas?

Is this the verray mede of your beheste?

Is al this peynted proces seyd, allas!

Right for this fyn? O lady myn, Pallas!

Thou in this dredful cas for me purveye;

For so astoned am I that I deye!

(Chaucer)

But suggestion can equally properly be the activation of one unspoken meaning instead of several:

No motion has she now, no force;

She neither hears nor sees;

Rolled round in earth's diurnal course,

With rocks, and stones, and trees.

(Wordsworth, 'A slumber did my spirit seal')

The unspoken meaning can often be a mood, and suggestion then would consist in evoking the emotion through its cor-relative sensuous details:

And from the first grey wakening we have found

No refuge from the skirmishing fine rain

And the wind that made the bell-tents heave and flap

And the taut wet guy ropes ravel out and snap.

All day the rain has glided, wave and mist and dream,

Drenching the gorse and heather, a gossamer stream

Too light to move the acorns that suddenly

Snatched from their cups by the wild southwesterly

Pattered against the tent and our up-turned dreaming faces.

(Alun Lewis, 'All day it has rained')

SUGGESTION TODAY

5

Suggestion is, most importantly, the diffusion of manifold meaning from a symbol or a system of symbols:

And all shall be well and

All manner of thing shall be well

When the tongues of flame are in-folded

Into the crowned knot of fire

And the fire and the rose are one.

(Eliot, ‘Little Gidding’)

Where suggestion is not the work of imagery, it can be a function of rhythm, of rhythm pointing to a meaning-in the following case, to a sense of the spirit’s release:

From the wide window towards the granite shore

The white sails still fly seaward, seaward flying

Unbroken wings.

(Eliot, ‘Ash Wednesday’)

Rhythm can sometimes be suggestive not in the sense that it suggests a particular meaning but in the sense that it is unresolved, disturbing, outward-pointing:

It was not the dark filling my eyes

And mouth appalled me; not even the drip

Of rain like blood from the one tree

Weather-tortured. It was the dark

Silting the veins of that sick man

I left stranded upon the vast

And lonely shore of his bleak bed.

(R. S. Thomas, ‘Evans’)

To make the same point negatively, suggestion can never

6

SUGGESTION TODAY

issue from the kind of rhythm that ticks with a self-contained

unbroken regularity:

The glories of our blood and state,

are shadows, not substantial things,

There is no armour against Fate,

Death lays his icy hand on Kings,

Scepter and Crown,

Must tumble down,

And in the dust be equal made,

With the poor crooked sithe and spade.

(Shirley, ‘Death the Leveller’)

The only justification for this incomplete catalogue of the

elements of suggestion, reading like a schoolboy’s answer to

the question, ‘What is suggestion?’, is that it could be useful

as a first quick look at the concept. Any full-length examina-

tion of the concept, however, involves an initial examination

of the content and status of the term, and here we come

across an ironical situation. Suggestion is clearly the pre-

vailing mode of modern poetry, and the exploration of

suggested meaning is the largest single concern of present-

day criticism and scholarship. One would therefore expect

‘suggestion’ to be one of the overworked literary terms of

our times, a cliché of criticism encountered on every page.

This, however, is hardly the case. The word is not even

found in the many glossaries of literary terms that have

appeared in recent years. With Edgar Allan Poe and the

Symbolists who were engaged in a struggle to establish

the case for suggestion in opposition to other values, the

word was a crucial term to be used systematically, pointedly,

maybe even polemically. But when the concept has met

with general acceptance and ceased to be a cause, the ascend-

ancy and currency of the term are, paradoxically, reduced.

SUGGESTION TODAY

7

The experience having attained universality, its name becomes generic, and the specializations that are now understandably developed are given their own names, and these become vogue words. Refinements of the concept of suggestion have thus been worked out and labelled, and their specific names—Emotive Meaning, Ambiguity, Obliquity, Irony/Paradox, Connotation, Intension, Qualitative Progression, Texture, Gesture and so forth—have tended to eclipse the generic name in prestige and popularity.

To eclipse it, yes, but not to eliminate it. Among the passages quoted earlier as specimens of suggestion are three that were selected because they have been analysed and commented upon by the inventors of new names for suggestion, and the language of their comment is significant. The lines from Troilus and Criseyde are examined in Empson’s chapter on the second type of ambiguity. ‘Ambiguity’, as used by him, is clearly a substitute for ‘suggestion’. And in a passage exploring the ambiguities—or more precisely, the puns—in the lines, Empson uses the word ‘suggest’ as many as five times:

. . . And rising behind that again, heard in the indignation of the phrase, is a threat that she may expose him, and peynted and fyn things?’ the reader may ask; and there is no obvious reply. It depends how carefully the passage is supposed to be read . . . It is a more crucial question how far peynted, in a proper setting, can suggest ‘pains’; how far we ought to leave the comparatively safe ground of ambiguity to examine latent puns . . . I have sometimes wondered whether Swinburne’s Dolores gets any of its energy from the way the word Spain, suggested by the title and by various things in the course of the poem, although one is forced to wonder what the next rhyme is going to be, never appears

B

8

SUGGESTION TODAY

among the dozen that are paired off with Our Lady of Pain . . . I

want to back up my 'pains' from peynted by calling in 'weighted'

and 'fainted' and the suggestion of labour in all that painted.1

The Wordsworth passage I have quoted was offered by

Cleanth Brooks as an example of 'Irony', a term he obviously

intends to be a replacement for 'suggestion'. The irony in

the passage, on Cleanth Brooks's showing, springs from the

opposition between the image of Lucy motionless in death

and the image of the earth's rotation which involves her.

But mark the phraseology of his exposition of the irony:

Wordsworth . . . attempts to suggest something of the lover's

agonized shock at the loved one's present lack of motion—of his

response to her utter and horrible inertness. And how shall he

suggest this? He chooses to suggest it, not by saying that she lies as

quiet as marble or as a lump of clay; on the contrary, he attempts

to suggest it by imagining her in violent motion—violent, but

imposed motion . . . (Italics mine).2

Tillyard similarly developed 'Obliquity' as a term to be

preferred to 'suggestion'. Yet in examining the rhythm of

Shirley's 'Glories of our blood' stanza as a case of the

absence of Obliquity, what Tillyard does say is: 'The

general tone of the metre, varied though it is by the short

couplet within each stanza, is that of enunciation, not of

suggestion.'3

Clearly the new terms cannot put 'suggestion' out of

business. Perhaps the older word is the better word, after all.

Also, as William Righter has argued, the terms embody

valuable insights but not precise explanatory concepts and

are not to be taken seriously as additions to the technical

vocabulary of criticism.4 What they have accomplished, I

think, is to provide 'suggestion' with an enlarged and

SUGGESTION TODAY

9

diversified content, now lending better definition to an

existing aspect, now adding a new one; and because of them

we have today fuller information than before about the

mechanics of suggestion.

Take, for instance, the oldest of the new terms—Emotive

Meaning (or Emotive Language). It was intended, when

Richards first brought it into use, as an umbrella term for the

many uses of language aside from literal or true-or-false

statement—as another term, in fact, for suggestiveness. It

renders explicit the most significant feature of suggestion:

the inter-relatedness of the suggestive and the emotive.

Richards has since moved so far away from his original

position as to claim that richness (or ‘resourcefulness’, as he

calls it) of meaning is not limited to emotive uses and is—

except for the very special category of mutually defining

technical terms—indeed universal in language. He takes

care, however, to add that ambiguity, while it exists every-

where, is in particular ‘the indispensable means of most of

our most important utterances—especially in Poetry and

Religion’.5 Suggestion and emotion would thus appear to

have a special relevance to each other after all.

‘... Any verbal nuance, however slight, which gives room

for alternative reactions to the same piece of language.’6

Here is what one would consider an excellent definition of

suggestion. It happens, however, to be Empson’s definition

of ‘Ambiguity’. His Seven Types is in fact the prototypal

study, in English criticism, of suggested meaning in poetry—

seeing suggestion not as aura or atmosphere but as the

presence of several distinct possibilities of meaning. Viewed

as such, suggestion is synonymous with Ambiguity and, of

all the new terms we have been discussing, Empson’s gave

most promise of establishing itself as the word for suggested

meaning. However, Ambiguity's actual achievement has only been to add a new possible facet to suggestion. Empson's fourth and seventh types represent linguistic ambiguity in a poem as an unconscious product of some psychological ambivalence in the poet—here certainly is something that could give the study of suggestion a new and exciting direction. The New Critics and others who followed Empson copied his minute inspection of the text but stigmatized any attention to the poet's personality as the 'Intentional Fallacy', so that the latter continues to be an interest of much potential value.

Tillyard's 'Obliquity'—reminiscent of 'Indirection' which is very senior, traceable in fact to Emerson and Whitman—is obviously yet another of the proprietary names for suggestion. At once the most rewarding and the most provocative part of his exposition of Obliquity is the notion that the oblique or suggested meaning should be alien, at least apparently alien, to the statement. His identification of the main indirect meaning of Lycidas as the Gīta's teaching, karmanyevādhikāraste Māphaleṣukadācana ('Action alone is thy province, never the fruits thereof'), is a bold act of interpretation. His claim that this piece of obliquity is properly validated despite the absence of overt support in the poem's statement makes it bolder still.7 There is, of course, the danger that if the link between what is said and what is suggested is too slender, the understanding will not be able to move from the one to the other. But the worse danger to the poem surely is from a hypertrophied, a tediously elaborate and obvious, link—like the correspondence between Mariana's mood and the too predictable imagery in Tennyson's poem. As Wimsatt has pointed out, too many deaths in Romantic poetry take place in winter or at

night and too few lovers meet except in spring or in the countryside.8 Tillyard's work on Obliquity has added

to the content of 'suggestion' the very useful idea of an optimum distance between statement and suggested meaning.

In Cleanth Brooks's use, 'Irony' seems to mean the action of the context in disrupting the dictionary meanings of

words and stimulating interinanimation between them; he also uses it to mean the association of discordant or con-

tradictory elements. In either aspect, Irony is identical with suggestion, and both aspects are equally significant, but in

Cleanth Brooks's hands the latter does seem to have been the source of a new emphasis that has helped in giving

suggestion its specifically modern interest. The yoking of incompatibles is nothing new and can be traced to Coleridge's

criticism or, farther back, to Aristotle's. but it is the technology of modern imagery that has made it the unique secret of

suggestiveness. To Cleanth Brooks, the value of juxtaposing non-congruent concepts or images is that it is an insurance

against sentimentality and a guarantee of honest taking note of the complexities and contradictions in experience. Its

value, however, to suggestiveness lies in the fact that the more disparate the objects you pull toward each other, the

greater seems to be the semantic energy released between them. For suggestion seen thus—as the bringing together of

disparates—F. W. Bateson has his own term, 'Semantic Synthesis'.

As well as being a term in logic where it means something altogether different, 'Connotation' stands for whatever a

word conveys other than its primary or dictionary meaning —other than, that is, its Denotation. This is how it is in

general use; in criticism, it can refer to musical evocations,

12

SUGGESTION TODAY

deposits of racial memory, traditional associations within or across cultures, private associations, colloquial overtones, implied ideas, affective colouring—in criticism, in short, ‘Con-notation’ has acquired interchangeability with ‘suggestion’. The value of the word lies in the very important proviso contained in the prefix con-. ‘Con-’ is ‘together with’; suggested meaning is what a piece of language carries together with (never without) its stated meaning—there can be no connotation without denotation. Those who would abolish discourse and attempt a rarefied language where the logical meaning of words is suspended and words are purely musical notes or magic devices or pieces of string must heed the voice of the little Latin morpheme, ‘con-’. Its message is, to borrow T. S. Eliot’s words, ‘ . . . You cannot have the aura alone.’9 Yvor Winters’s—or Kenneth Burke’s—term ‘Qualitative Progression’, which is used pejoratively, refers to the kind of suggestiveness which offers a series of images that may have internal coherence but are not buttressed by a structure of narration.

‘Connotation’ and ‘Intension’ mean roughly the same in logic, and Allen Tate’s adoption of ‘Intension’ as a substitute for ‘suggestion’ looks like a case of this near-synonymity getting duplicated in criticism—although the explanation for his being attracted to the word lies less perhaps in logic than in metaphysics. Bergson’s ‘Intensive Manifold’, which, because of Hulme, has been a seminal concept in modern criticism, has to do with intuition, unanalysable wholeness and imperviousness to paraphrase, and although Tate’s own method of exploring ‘intensive meaning’ is verbal analysis, I think his notion of ‘Intension’ (which is part of a theory of ‘tension’) preserves, more than vestigially, the anti-intellectual orientation of Bergson’s concept. The service

that Tate's term does to criticism is to bring into focus the basically extra-logical character of suggestion.

John Crowe Ransom's term 'texture' denotes a poem's concrete sensuous features—the sound values, the imagery, the connotations of its language—which establish its poetic non-prosaic identity. This is, as he sees it, one component of the poem, the other being the paraphrasable statement in it which he calls 'logical structure'. This seems to be the old statement-suggestion antithesis in a different wrapping, and in so far as it is possible to discern any concepts through Ransom's somewhat opaque exposition, it does seem likely that 'texture' represents a struggle to find a more precise and expressive word for suggestion. The principal gain from Ransom's restatement is the pointing up of 'concreteness' in the suggestors as vital to a poem per se.

It is not easy to see quite what Blackmur means by 'Gesture', chiefly because the language in which he expounds the meaning of the term shares some of that tendency to advance beyond the literal which he regards as the differentia of language as gesture. The examples are even less helpful than the explanations. There are, however, one or two revealing asides: '. . . that play of meaningfulness among words which cannot be defined in the formulas in the dictionary, but which is defined in their use together'10; '. . . the revelation of the sum or product of all the meanings possible within the focus of the words played upon . . .'11 It is difficult to resist the feeling that 'Gesture' so conceived is suggestion in a new suit. A valuable insight that Blackmur's essay contributes to the theory of suggestion concerns the difference between Gesture and symbol. 'Gestures are the first steps toward the making of symbols, and those symbols which endure are the residuary legatees of the meanings

14

SUGGESTION TODAY

earned through gesture.'12 The distinction between a symbol accomplished and a quantity of suggested meaning that is yet to develop permanence or recurrence and graduate to the status of symbol—the distinction, in other words, between suggestion as being and suggestion as becoming—can be an important one, enabling us to see suggestion as a two-stage process.

Profiting by competition with younger and more prestigious rivals, 'suggestion' seems to have drawn varied sustenance from them and gained more definition and improved differentiation. But enrichment has come to the word in another way too—not at the expense of new equivalents competing with it, but by being itself adopted as the English equivalent of an ancient Sanskrit term. The first Indologist to translate dhvani as 'suggestion' must have done so not because he was concerned to establish parallels between a ninth-century Indian critical tradition and a nineteenth-century Western one but merely because it seemed to be the obvious equivalent in English for dhvani or vyañjanā. The literal meaning of dhvani is 'sound'. (It is interesting that the concept of suggestion is expressed in English at times by means of sonic terms like resonance, reverberation, overtones and undertones.) Dhvani was originally a term in linguistics where it referred to the final sound in a word, which, when apprehended, suggested or revealed the phonological identity of the whole word. Borrowing the word and making it stand for the suggestion of meaning, Sanskrit semasiology and poetics had evolved, by about the middle of the ninth century, a theory of poetic meaning constructed round the notion that suggestion is a distinct function of language and indeed the principle of the highest kind of poetry, and had offered, in Ānandavardhana's Dhvanyāloka, a complete for-

mulation of the theory. The theory grew out of a body of

poetry very different from the output of nineteenth-century

romanticism, and the modern European languages in which

this movement flowered are a far cry from Sanskrit; also,

the methodology of Ānandavardhana and his successors,

which was ruled largely by deduction, definition and classi-

fication, has little in common with the modus operandi of the

Symbolist critics or even of the more cerebral New Critics.

Even if these divergences were not there, one would still

regard analogies assumed between cultures remote from each

other as fancied or superficial. Nevertheless, it is clear that

in a very real sense the exponents of dhvani and the ex-

ponents of suggestion have been looking at the same

phenomena and reaching the same findings—such as, the

importance in poetry of what is not stated, the many-

levelledness of poetic meaning, the alogical nature of all

apprehension of unstated meaning in poetry, and objectifica-

tion as the only mode of presenting emotion in poetry. The

affinities ring so true that it is impossible for anyone aware

of both systems to use the word ‘suggestion’ without think-

ing of all that dhvani denotes. The fertilization of suggestion

with notions from Sanskrit is a real enough event—probably

the most significant thing that has happened to the concept

since the Symbolists provided Poe’s airy-fairy formulation

with enough conceptual meat. If this accession of meaning

to ‘suggestion’ from Sanskrit thought has not been in-

fluential, it is only because the available expositions in

English of the dhvani theory have either not been sufficiently

comprehensible or have failed to reach more than a small

circle of readers in the West.

The status of ‘suggestion’ is bound up with that of its

opposite, ‘statement’, which has a shorter history—Poe

16

SUGGESTION TODAY

preferred the phrase 'the upper current of meaning' and

Mallarmé preferred nommer and description. The growth of

'statement' as a critical term has been since the turn of the

century. More recently, it has had to work in competition

with 'discourse', 'narration', 'argument', 'extension', 'de-

notation', 'Rational Progression' and other near-equivalents.

'Statement' and 'suggestion' have, however, persisted as a

pair of antonyms with more serviceability and authority

than any other. And unlike 'statement' which denotes what

is more the exception than the rule in contemporary poetry,

'suggestion' has been acquiring greater precision as well as

greater range as a generic naming device and developing ex-

tensive descriptive and evaluative uses. And, whether named

as such or not, the notion itself, favoured by an aesthetic

which makes the rejection of verbal explicitness the very

condition of artistic expression, is now established as a key

concept and as a versatile means of analysis and interpreta-

tion and is constantly adding to its refinements and diversi-

fying and enlarging its uses.

2

Suggestion: From Poe to the

Present

The first observed use of the word ‘suggest’ in what seems to

come closest to its present sense in criticism is in Dryden’s

sentence (1697), ‘Virgil . . . loves to suggest a truth in-

directly’.1 Another early occurrence is in Lord Kames’s

Elements of Criticism (1785): ‘ . . . and by suggesting various

meanings at once, it [viz. “a vague or obscure expression”] is

admired by others as concise and comprehensive’.2 Dryden

and Kames clearly are using the word here in its popular

sense (viz. ‘suggestion’ as the opposite of ‘direct expression’)

while making a critical statement. Which is, of course, very

different from using it as a critical term, as for instance

David Perkins does today when he says: ‘No more than

other romantic writers does Wordsworth spend much time

speculating how the process of suggestion actually works.’3

In any case, the golden age of English statement poetry, to

which Dryden and Kames belonged, can hardly be expected

to have witnessed the hypostatization of suggestion. A more

likely season for this would be the nineteenth century in

England and Europe. Yet ‘suggestion’, as far as I am aware,

does not occur as a technical word anywhere in Romantic

criticism. In On the Constitution of Church and State According

to the Idea of Each, Coleridge does speak of ‘ideas which may

indeed be suggested and awakened, but cannot, like the

17

18 SUGGESTION: FROM POE TO THE PRESENT

images of sense and the conceptions of the understanding,

be adequately expressed by words'. But the coupling of

'suggested' with 'awakened' shows that 'suggested' is not

a technical word here any more than 'awakened' can be, and

that Coleridge is merely using the word here like Dryden

and Kames in its popular sense as the opposite of 'directly

expressed'. Nor is the word, as far as I am aware, used in a

technical sense in Goethe, Schiller, Schelling, the Schlegels

and the other German Romantic critics. Indeed, Edgar Allan

Poe, while crediting the German critics with the concept of

a suggestive under-current of meaning in poetry, points out

that their word for poetry with this under-current is 'mystic'.

Another area where one may hope to locate the origin

of 'suggestion' as a technical term is American criticism.

Although early American culture—both the commercial

practicality of its material aspect and the austere Puritan

plainness and logicality implicit in its spiritual aspect—was

firmly wedded to the concept of language as unambiguous

rational discourse, yet basically the New England imagina-

tion was allegorical or even—however incipiently—symbol-

istic, and we find Cotton Mather, a younger contemporary

of Dryden, moving so far away from his earlier commit-

ment to univocal simplicity as to approve of allusiveness

and implication. In the eighth chapter of Manuductio ad

Ministerium (1726), he commends the style where 'the

paragraph is embellished with profitable references, even to

something beyond what is directly spoken. Formal and pain-

ful quotations are not studied, yet all that could be learnt

from them is insinuated.'4 This comes pretty close to sug-

gestion, but Mather nowhere uses the term, nor do any of

the other American rhetoricians and critics who preceded

Poe.

SUGGESTION: FROM POE TO THE PRESENT

19

A few scattered instances of the use of 'suggest' and 'suggestive' can indeed be met with in critical writings from Dryden onwards, but clearly in none of these cases is the word used as a critical term. When does a popular word become a technical term? Of course, purely formal devices like italics, quotation marks or an initial capital letter can signal the intention that the word is technical, but in substance a word is raised above the lay lexis and established as a technical term when an element in its original popular meaning is selected and fixed by explicit definition and conscious and consistent specialized use. By this token, it is, I think, in Poe's critical writings that 'suggestiveness' is first used as a technical term.

During the years 1840–46 the nature of the imagination and of suggestion seems to have engaged Poe's attention a great deal. In his review of Thomas Moore's poem Alciphron, Poe rejects Coleridge's famous definition of Imagination and Fancy and sets up his own. Poems 'which mankind have been accustomed to designate as imaginative' are remarkable, he says, for their 'suggestive character', by which is meant that 'there lies beneath the transparent upper current of meaning an under or suggestive one'. This 'mystic or secondary expression' of sentiment 'has the vast force of an accompaniment in music'. 'With each note of the lyre is heard a ghostly, and not always a distinct, but an august and soul-exalting echo . . . . But not so in poems which the world has always persisted in terming fanciful. Here the upper current is often exceedingly brilliant and beautiful; but then men feel that this upper current is all.'

This is a clear enough formulation (although the terminology makes it sound somewhat inchoate) of the distinction between suggestive poetry and statement poetry.

20 SUGGESTION: FROM POE TO THE PRESENT

But what is significant here is that in Poe's use the terms 'suggestive' and 'imaginative' are so intimately related as to be interchangeable. To find out the nature of the connection, we have to turn to The Poetic Principle. Here he defines the end of poetry as 'the Rhythmical Creation of Beauty'—beauty 'that appertains to eternity alone'—by means of the imagination; and he implies that the imagination works through 'multiform combinations among the things and thoughts of Time'. 'It is in Music, perhaps', he goes on in the same essay, 'that the soul most nearly attains the great end for which, when inspired by the Poetic Sentiment, it struggles—the creation of supernal Beauty . . . And thus there can be little doubt that in the union of Poetry with Music in its popular sense, we shall find the widest field for the Poetic development.' What makes music the vehicle par excellence of the sense of Ideal Beauty? Poe had provided the answer several years earlier as a young man of twenty-two in the 'Letter to B—' (published as a preface in his Poems, 1831), where he spoke of poetry 'presenting perceptible images . . . with indefinite sensations, to which end music is an essential, since the comprehension of sweet sound is our most indefinite conception'. Indefiniteness is essentially suggestive. It is then by indeterminateness of expression (a quality that music possesses in a pre-eminent degree) that in poetry the imagination suggests 'Beauty'. This assumption explains why what is imaginative is invariably suggestive and why poetry is most itself when it shares the nature of music. The assumption is elaborated in Marginalia (XVI, 29): '. . . a suggestive indefinitiveness of meaning, with the view of bringing about a definitiveness of vague and therefore of spiritual effect . . . I know that indefinitiveness is an element of the true music—I mean of the true musical expression. Give to it any

SUGGESTION: FROM POE TO THE PRESENT

21

undue decision--imbue it with any very determinate tone--and you deprive it, at once, of its ethereal, its ideal, its intrinsic and essential character. You dispel its luxury of dream. You dissolve the atmosphere of the mystic upon which it floats. You exhaust it of its breath of faery. It now becomes a tangible and easily appreciable idea--a thing of the earth, earthy.

This explains why Poe, while he always uses the term 'suggestive' with a precise meaning, also uses it inclusively, equating it with 'imaginative' and often with 'mystical', 'ideal', 'beautiful', 'ethereal', 'spiritual', 'elevating', 'august' and 'pure'.

The concept of suggestiveness, stated in such very general terms, looks fairly convincing. But when Poe comes down to brass tacks, the effect is anticlimactic. The poems he enumerates as remarkable for their suggestiveness--Prometheus Vinctus, The Inferno, The Ancient Mariner, Christabel, Kubla Khan, The Nightingale Ode, The Sensitive Plant, etc.--are perhaps suggestive in varying degrees, but they certainly do not isolate the differentia of suggestive writing. Poe calls our attention to 'the suggestive force which exalts and etherealizes' the passages he quotes from Alciphron, but anyone who looked for such force in the lines would be looking in vain. Nor would many readily join Poe in rating Tennyson 'the noblest of poets' for suggestiveness.

The metaphors Poe employs to expound suggestiveness --'a ghostly and not always a distinct . . . echo'; 'long and wild vistas'; 'dim bewildering visions'--seem to point to the symbol's freedom from equivalence to a concept and its capacity for resonant multiple meaning. Yet in Poe's exposition of his own poem 'The Raven', it turns out that the 'under-current, however indefinite, of meaning' from which springs the suggestiveness of the last two stanzas is, after all,

22

SUGGESTION: FROM POE TO THE PRESENT

merely the idea of 'mournful and never-ending remembrance' which the bird stands for. Similarly (although we do not have Poe's word for it) the palace in 'The Haunted Palace' can stand for a deranged mind, and 'the spectre of a planet' in 'Ulalume' can stand for the semblance of love. But this is a paradox that plagued French Symbolism as well. It in fact inheres in the art of suggestion—a poet who knows that he is being suggestive cannot help suggesting something; a consciously used symbol inevitably develops reference. But Poe shows a way out of the dilemma. In his review of Hawthorne's Twice-Told Tales and Mosses from an Old Manse, it turns out that by 'the suggested meaning' Poe means the allegory, but it is the allegory 'seen only as a shadow or by suggestive glimpses'. The thing symbolized should never be evident or explicit. He seems to make the same point in The Philosophy of Composition: 'It is the excess of the suggested meaning—it is the rendering this the upper instead of the under current of the theme—which turns into prose (and that of the very flattest kind) the so called poetry of the so called transcendentalists.'

Poe's concept of indefiniteness is very suspect today—we have moved away from it and all the way along the scale to Hulme's concept of 'accurate, precise and definite description'. Also, the words Poe uses as near-synonyms of 'suggestive'—words like 'ideal', 'ethereal', 'elevating', 'august', and 'pure'—will offend present-day taste. In fact Poe's whole formulation of the principle of suggestiveness is in terms that tend to validate second-rate Romantic poetry.

But it was Poe's broad theoretical formulation broadly apprehended, and not at all either his specific applications of it or his own poetic practice, that made 'suggestion' a key

term in subsequent criticism. The consistent and self-conscious, if a little immature and ebullient, use of the word 'suggestive' in Poe's incidental theorizing in successive essays and reviews had the effect of establishing it as a technical term in criticism. It was transmitted from Poe to his younger contemporaries and successors in American criticism. Henry Timrod's essay, Theory of Poetry (1863), which is largely a defence of the long poem against Poe's attack, reaffirms nevertheless Poe's principle of suggestion and does so in language very similar to Poe's:

... and the poetry of words has never so strange a fascination as when it seems to suggest more than it utters, to call up by implication rather than by expression those thoughts which refuse to be embodied in language, and to hint at something ineffable and mysterious of which the mind can attain but partial glimpses.5

Both Emerson and Whitman use 'indirections' as an alternative term to 'suggestion'. Whitman, however, uses the term 'suggestion' at least twice. In 'Democratic Vistas' he mentions how the poet 'seldomer tells a thing than suggests or necessitates it'. In 'A Backward Glance O'er Travel'd Roads' (1888), the preface to the ninth and last edition of Leaves of Grass, he names the distinctive quality of the Leaves: 'The word I myself put primarily for the description of them as they stand at last is the word Suggestiveness.'

It was French criticism, not American, that seriously took over the concept of suggestiveness, developing full-bloodedly the various significations it had acquired in Poe's handling of it and erecting a poetic round it. Poe's triple sway over nineteenth-century French literature—as man, as poet and

C

24 SUGGESTION: FROM POE TO THE PRESENT

as critic—is one of the strangest episodes of literary history. Baudelaire, Mallarmé and Valéry, although they essentially formed a single tradition, received Poe's influence independently of each other and through an imperfectly learnt English language. J. Isaacs felt that in particular Poe's influence on their criticism—specifically, in regard to the notion that poetry partakes of the nature of music—was uncanny, as none of them, he was sure, could have read Poe's review of Alciphron where this notion is offered.6 But it is also offered, as our quotations above show, in his other critical writings which were accessible enough to his French admirers, and there is no need to suspect clairvoyance.

The French Symbolists' theoretical pronouncements seem themselves to exemplify some of the virtues—indirection, vagueness, obscurity—that they celebrate as the soul of poetry, so that in making a summary in order to show how the concept of suggestion can be claimed to have generated the tenets of Symbolism, it is necessary to reduce everything to points even if that tends to make things appear clearer and simpler than such things ever could be.

1. Objects in the phenomenal universe

(a) suggest the supra-sensible reality beyond them, and

(b) suggest one another across the barriers between the arts and across the barriers between the senses. (‘Ce qui serait vraiment surprenant, c'est que le son ne pût suggérer la couleur, que les couleurs ne pussent pas donner l'idée d'une mélodie, et que le son et la couleur fussent impropres à traduire les idées.' Baudelaire)

2. The poet uses these objects

(a) to suggest macrocosm (i.e., transcendental reality/ the Idea/the Essence), or

(b) to suggest microcosm (i.e., a state of mind, an

SUGGESTION: FROM POE TO THE PRESENT

25

emotion). (‘ . . . choisir un objet et en dégager un état d’âme,

par une série de déchiffrements.’ Mallarmé)

Neither of these would admit of being presented otherwise

than by suggestion.

3. The poet presents the object itself not by naming or

description but by suggestion (i.e., allusion, gradual evoca-

tion). (‘Nommer un objet, c’est supprimer les trois quarts de la

jouissance du poème qui est faite de deviner peu à peu: le suggérer,

voilà le rêve. C’est le parfait usage de ce mystère qui constitue le

symbole: évoquer petit à petit un objet . . .’ Mallarmé)

4. The poet’s word instead of representing the sensuous

exterior of the object (the palpable trees) suggests the essence,

the atmosphere (the forest’s shudder).

5. The poet abolishes the word’s meaning (= logical

meaning) and liberates suggestion (= associations) from its

sound.

6. Music is the suggestive art par excellence, performing all

the functions listed above better than any other art. Poetry

should strive to approximate to music. (‘Et pour suggérer les

émotions . . . un signe special a été inventé: le son musical.’

Wyzéwa)

7. Indefiniteness/mystery/obscurity as elements of sug-

gestiveness. (‘. . . quelque chose d’un peu vague, laissant carrière

à la conjecture . . . Le mystère, le regret sont aussi des caractères

du Beau.’ Baudelaire)

These jottings will serve to show how suggestion, ex-

plicitly so called, is central to the Symbolist aesthetic; it

would be strange if it were not, for the symbol is after all

a suggestive device. It will also be clear that the main

features of this carefully wrought aesthetic can be traced

to the insights that Poe expressed so fumblingly and super-

ficially in his theorizings on suggestion: Ideal Beauty;

26

SUGGESTION: FROM POE TO THE PRESENT

indefiniteness; poetry sharing the nature of music; and

(although I have not touched upon these) suggestiveness

implying short-flightedness and the-poem-for-its-own-sake.

The noblest tribute Poe has earned so far from anywhere

comes from that doyen of Australian literature, A. D. Hope:

'There would be no point in contemplating the ingenious

but essentially trivial arguments of Poe's The Poetic Principle

had these arguments not had the misfortune to justify so

aptly what was going on in poetic practice in the nineteenth

century, and had they not, by a series of unfortunate

accidents, become the basis of so many varieties of modern

poetic doctrine.'7 The biggest of these unfortunate accidents

is, of course, the Symbolist Movement.

Any survey of the history of suggestion in our own century

is complicated by the paradox (mentioned in the last chapter)

that while the phenomenon has become universal in litera-

ture and interest in it has become universal in criticism, the

term itself has limited currency, so that the historian has to

restrict his range to suggestion-named-as-such. This is a

large enough area though, and within it the concept, while

preserving a recognizable continuity from Poe to Perkins,

has moved along new lines of development.

At the turn of the century, A. C. Bradley, in his Oxford

Inaugural, Poetry for Poetry's Sake (1901), described sug-

gestion as a case of the poem pointing beyond itself to an

all-embracing perfection which can only be partially mani-

fested through language; by a mysterious resonance, what

is stated expands into an infinity of suggestion.8 Bradley

was the first heir to the spiritual-transcendental conception

of suggestion that the Symbolists worked on; he was probably

also the last, as attention was soon to shift from the magical

SUGGESTION: FROM POE TO THE PRESENT

27

to the semantic aspect of suggestion. In philosophy, in linguistics, in literary criticism, the new century was one that was destined to become a uniquely semasiological era. Already, a year before Bradley's Inaugural, Yeats in the essay The Symbolism of Poetry (1900) had spoken of 'innumerable meanings, which are held to "white" or to "purple" by bonds of subtle suggestion.'9 This is not to say that interest in suggestion as a mystique died out altogether. In his Phases of English Poetry, Sir Herbert Read is found saying: 'Suggestion, in my opinion, covers all those vague notions which the Abbé Bremond has wrapped in fluffy phrases, such as "mystery", "enchantment", "intimate nature of the soul", "magic" and so on.'10 This was in 1928, but about this time suggestion was already parting company with vagueness and fluffiness.

The transition from suggestion-as-magic to suggestion-as-meaning is vividly demonstrated in Lascelles Abercrombie's critical ambivalence. He could speak of 'verbal magic' and 'the enchantment of language'11 but he also defines suggestion as the business of liberating the secondary meanings -intensities and complexities of sense-that cluster round the word's dictionary definition, in order to render the 'sensuous impressions, psychological intuitions and the mass of infinitely variable associations' that, along with the thought element they surround, compose the experience the poet is trying to express.12

One of the earliest to question the Symbolist assumption that meaning and suggestion are opposed principles was, interestingly enough, a distinguished heir to Symbolism: T. S. Eliot. In the essays on Marvell and Dryden (1921) he notes how in Morris's poetry (or at least in the lines he examines) and in Swinburne's, suggestion is abortive because

28 SUGGESTION: FROM POE TO THE PRESENT

statement is absent—their language is all connotation and no denotation.13 By replacing suggestion-versus-meaning with the antithesis connotation-versus-denotation, Eliot equates suggestion with connotational meaning. The demolition of the theory of suggestion as non-meaning was completed by F. W. Bateson in his English Poetry: a Critical Introduction (1950) with the dictum, ‘A word’s connotation is as much a part of its meaning as its denotation’.14 Four years later, in The Verbal Icon, W. K. Wimsatt was even claiming that ‘suggestion’ was a better word than ‘connotation’ for this department of meaning; the suggestion/statement antithesis, he felt, was simpler and clearer, as connotation/denotation and intension/extension have a different technical sense in logic and can thus be misleading.15 A more recent description of suggestion (by John Bayley in The Romantic Survival, 1957) as ‘casting the net of reference in a wider and more subjective arc’16 shows that the process of redefining suggestion as a branch of meaning is now complete.

Parallel to this growing attention to suggestion as meaning there has been the development of interest in suggestion as an event in the reader’s consciousness. Defining suggestion negatively in terms of the Parnassians’ method, Mallarmé had said of them: ‘. . . ils retirent aux esprits cette joie délicieuse de croire qu’ils créent.’ And in the passage on suggestiveness (which I quoted from) in Whitman’s preface to his Leaves of Grass, he further says:

I round and finish little, if anything; and could not, consistently with my scheme. The reader will always have his or her part to do, just as much as I have had mine. I seek less to state or display any theme or thought, and more to bring you, reader, into the atmosphere of the theme or thought—there to pursue your own flight.

SUGGESTION: FROM POE TO THE PRESENT

29

This approach to suggestion, which regards suggested meaning as a construct of the reader’s mind, has had greater appeal for some than suggestion as the harvest of the text. In Poetic Diction (1928), Owen Barfield makes the point that meaning essentially is intuited by the listener or reader and that suggestion is the play of ‘poetic imaginative sympathy’ which makes such intuition possible.17 The theory of suggestion as reader-participation has been an abiding interest at Harvard. W. J. Bate understands by suggestion the way poetry, through indefiniteness or incompleteness, rouses the reader’s imagination to creative activity.18 His student, David Perkins, in his Wordsworth and the Poetry of Sincerity (1964), refers to the notion of suggestion as ‘the calling up of what is already there in the reader’s mind’, only to point out that this is not what happens when we read The Prelude. When mystic experience such as Wordsworth’s, to which the reader’s own life has no parallels, has to be communicated, suggestion operates by getting the reader’s mind to ‘work creatively upon the poem’, so that the reader ‘goes beyond what can be said in words, himself leaping the gap and arriving at the intuition in the poet’s mind’.19 The reader’s collaboration is, in fact, what distinguishes suggestion in literature from suggestion in cinema. In an Encounter article which appeared some eight years ago, Nicola Chiaromonte makes the point that the non-verbal suggestive effect of the cinematographic image is immediate and deterministic while the response to the suggestiveness of language calls for the mind’s active intervention: ‘The cinema derives its power from its ability to arouse an emotional reaction that is both immediate and certain. Whereas a poem or a novel cannot come alive without the reader’s elaboration; its power of suggestion is a construction of his mind, calling into

30

SUGGESTION: FROM POE TO THE PRESENT

play his sensibility, and his intellectual and imaginative faculties.'

Exciting as it can be, speculation is all that is possible about what happens in a mind exposed to suggestion.

Certainly, the more worthwhile of the two lines of development that the concept has followed is the one that has obeyed the trend in modern taste and criticism and led to the identification of suggestion with unstated interpretable layers of meaning.

It is this aspect of it that has absorbed the additions and refinements offered by Ambiguity, Irony and other novel formulations and acquired a depth, reach and precision that could more than make up for its not being a vogue word.

3

Suggestion through the

Objective Correlative

If people do not talk about the 'objective correlative' any more, it is not because it has been found untenable and furtively dropped. Rather it has ceased to be debatable—its axiomatic status has now been conceded. The criticisms brought up when T. S. Eliot enunciated it in 1919 were that the phrase was ugly and the idea one of strictly local validity at best, that it applied only to dramatic poetry, that it really applied only to Eliot's work, that it applied only to Hamlet, that it did not apply even to Hamlet. Eliot now faces objections which are the very opposite of these—that the idea is nothing new and that even the phrase (which thirty-eight years after he first used it Eliot was still feeling pleased with himself for having coined)1 is not Eliot's but one Washington Allston's.2 (Credit for 'dissociation of sensibility' has similarly been transferred by Mr Bateson to Rémy de Gourmont.3) But if Eliot is to have no share of the honours, the universality of the objective correlative has nevertheless now been accepted to the extent that the most influential criticism of our times is concerned with the intensive study of the correlates of emotion—imagery, sound values, and to a lesser extent, plot and character—within the verbal organization where they occur and belong.

31

32 SUGGESTION: OBJECTIVE CORRELATIVE

Eliot's critics are right—the objective correlative was neither invented nor discovered by Eliot. Indeed its value lies precisely in the fact that it was not invented or discovered by him. It is as old as poetry itself. If the impression has gained ground that Eliot in his criticism erected the objective correlative as a principle and kept talking about it while he was writing the kind of poetry which works by objectification of feeling through dramatic detail and that his criticism put the objective correlative aside when his poetry did not need it any more, he has largely himself to blame. In 1942 in his Ker Memorial lecture on The Music of Poetry, Eliot the elderly arbiter of taste went out of his way to cast doubt on the bona fides of Eliot the young critic. As a critic, the poet, he said, 'at the back of his mind, if not as his ostensible purpose, is always trying to defend the kind of poetry he is writing, or to formulate the kind that he wants to write'.

This is the case, Eliot warned his audience, 'especially when he is young, and actively engaged in battling for the kind of poetry which he practises'.4 Eliot, the young poet-critic of the 'twenties, by claiming for the method of myth and image the status of a discovery, had given the impression of setting up a theoretical formulation which would be an oblique apology for his poetry. When Ulysses appeared, he hailed its use of 'the mythical method'—the rendering of emotion or inward experience through direct sensuous embodiment in concrete image—as a revolution. It was less a revolution than a correction of the tendency to discursiveness that had been steadily growing as the mythopoeic faculty, after Keats and Shelley, weakened. But the various guides to T. S. Eliot took the cue from his manifesto for the mythical method and proceeded to show industriously how each poem of Eliot's was an old curiosity shop of objective cor-

relatives, from Mr Apollinax's head rolling under a chair or grinning over a screen right down to the twisted shape at the first turning of the second stair in Ash Wednesday. Here the guides paused, for Four Quartets had signally fewer of these oddments to offer. It was then assumed that Eliot had passed from dramatic objectification to direct poetic philosophizing. And so he had, but only fitfully, and as large parts of the Quartets show, the dramatic externals, the sensory correlates were never really dispensed with. As B asks in A Dialogue on Dramatic Poetry, what great poetry is not dramatic?5

The objective correlative as formulated by Eliot—palpable, precise, clear—was sired by the Imagist image. Hulme's orthodox doctrine, of course, regarded the dry, hard, specific 'thing' as self-sufficient, needing no emotional complement, but Pound diluted this to the extent of conceding that the natural object, while being its distinct concrete definite self, 'presents an intellectual and emotional complex'. But that, apart from the magie and the mysticisme, was what the Symbolist symbol was already doing. Mallarmé's salle d'ébène or his pourpre la roue du seul vesperal de mes chars is the external correlate of an otherwise ineffable état d'âme.

Objects as suggestive associates of emotion are necessarily present in all poetry, although one period, type or form may need them more than the others do or use them differently. It may be said—but only as a very general principle—that in art, states of sentience are expressed through their sensuous equivalents. As Wimsatt and Beardsley put it in their essay, 'The Affective Fallacy', emotions in poetry are 'not communicated to the reader like an infection or disease, not inflicted mechanically like a bullet or knife wound, not administered like a poison, not simply expressed as by

34 SUGGESTION: OBJECTIVE CORRELATIVE

expletives, grimaces or rhythms, but presented in their

objects and contemplated as a pattern of knowledge'.6

The locus classicus on the objective correlative is the latter

half of Eliot's essay on Hamlet:

Mr. Robertson is undoubtedly correct in concluding that the

essential emotion of the play is the feeling of a son towards a

guilty mother . . . The only way of expressing emotion in the

form of art is by finding an 'objective correlative'; in other

words, a set of objects, a situation, a chain of events which shall

be the formula of that particular emotion; such that when the

external facts, which must terminate in sensory experience, are

given, the emotion is immediately evoked . . . The artistic 'in-

evitability' lies in this complete adequacy of the external to the

emotion; and this is precisely what is deficient in Hamlet. Hamlet

(the man) is dominated by an emotion which is inexpressible,

because it is in excess of the facts as they appear . . . His disgust is

occasioned by his mother, but . . . his mother is not an adequate

equivalent for it . . . And it must be noticed that the very nature

of the données of the problem precludes objective equivalence. To

have heightened the criminality of Gertrude would have been

to provide the formula for a totally different emotion in Hamlet;

it is just because her character is so negative and insignificant that

she arouses in Hamlet the feeling which she is incapable of

representing.7

This could be Ānandavardhana in the ninth century

judging a play or a poem in terms of the theory of Rasa-

dhvani (the Suggestion of Emotion) of which he was the

first exponent. Eliot's concepts and terms here are uncannily

similar to those of Sanskrit criticism. The 'essential emotion'

is the dominant rasa. Disgust is jugupsā, one of the eight

sthāyins (permanent emotions). Gertrude is the human object

of the emotion—the ālambanavibhāva. The inadequacy of

objective equivalence that Eliot complains of would have been called a rasa-doṣa (a flaw in the presentation of emotion) by Mammata (11th–12th century) and an anaucitya (a case of inappropriateness) by Kṣemendra (11th century). The only difference (though it is an important difference) is that the Sanskrit terms denote a failure on the dramatist’s part, while Eliot’s point is that the inadequacy in objectification inheres in the situation in the play.

The cardinal concept of Indian aesthetics is rasa. Rasa literally means tincture, taste, flavour, relish. It has been variously translated as ‘Sentiment’, ‘Aesthetic Emotion’, Stimmung, Geschmack, Saveur. But all these words have other connotations which can mislead and confuse. What then is rasa? There is the well-known enunciation by Viśvanātha (14th century):

Rasa, experienced by men of sensibility, is born of the dominance of the sattva principle, is indivisible, self-manifested, compounded of joy and consciousness, untouched by aught else perceived, brother to the realization of brahman, and its very life is unearthly wonder.

This is so loftily metaphysical that one is glad to fall back on Bharata’s simple metaphor for rasa—the sensation on the palate! Even more simply, rasa can be described as the response to art. It has all the features of the aesthetic experience familiar to Western philosophy—it is emotion objectified, universalized; and raised to a state where it becomes the object of lucid disinterested contemplation and is transfigured into serene joy . . . This is as far as a non-philosopher can get in defining the nature of the rasa experience.

Rasa thus grows out of, is a consummation of, the emotion

36

SUGGESTION: OBJECTIVE CORRELATIVE

presented in art. Ancient Indian psychology—the one, at any rate, which Abhinavagupta (10th–11th century) made use of

—assumed nine heads under which all emotional activity could be grouped: the sexual emotion, amusement, distress,

anger, masterfulness (energy), fear, disgust, wonder and subsidence. These are permanent emotions (sthāyins) sprung

from universal psychic dispositions (vāsanās) in human nature and lie inert within a man except when a stimulus

activates them for a while. (Sthāyins are a universal human equipment, and all men are potentially capable of the

realization of rasa. Actually, however, only the sahrdaya or rasika—the man of either inborn or trained sensibility—

can realize rasa.) The nine categories are mutually exclusive and claimed to be together exhaustive, so that whichever

emotion the poem or play expresses or arouses must come under one sthāyin or the other.

How does a poem present or convey an emotion? Sanskrit theory has an answer that is by no means unique to

it: a poem does so through the objective correlatives of the emotion. Images, characters, situations which are the objec-

tive correlatives of the emotion are presented descriptively in a poem and, when the reader’s mind makes contact with

these, they awaken the corresponding sthāyin within him and raise it to the state of rasa. Nine specific rasas grow out

of the nine sthāyins:

Sthāyins

Rati (the sexual emotion)

Hāsa (laughter/amusement)

Śoka (grief/distress)

Krodha (anger)

Utsāha (masterfulness/energy)

Rasas

Śṛṅgāra (love)

Hāsya (the comic)

Karuna (pathos)

Raudra (anger)

Vīra (the heroic)

SUGGESTION: OBJECTIVE CORRELATIVE

37

Bhaya (fear)

Jugupsā (disgust)

Vismaya (wonder)

Śama (subsidence)

Bhayānaka (fear)

Bībhatsa (disgust)

Adbhuta (wonder)

Śānta (serenity)

The first three acts, to take an example, of Kālidāsa’s

Śākuntala (c. 4th century) are an idyll that fairly teems

with the objective correlatives of the śṛṅgāra rasa. There is

Śakuntalā herself, the hermit’s adopted daughter, who has

lit a blaze in the King’s heart. The creeper in bloom, the

mango tree with its young sprouts, the bold and greedy bee,

the bower on the bank of the Mālinī, the drowsy warmth

of the lotus-scented air, the uśīra paste spread to cool the

fever of Śakuntalā’s breast, and many more intoxicants of

the heart fill king and maid with an overmastering passion.

It will not be held down either by the timidity and bash-

fulness of the girl or by the circumspection and dignity

of the King and declares itself in tokens of word, look,

gesture and action—by the fever in Śakuntalā’s limbs, by

the colour that rises to her cheeks, by the distraughtness

and emaciation of the King. The love of Dusyanta and

Śakuntalā reaches the reader’s imagination through this

multitude of sensory correlates and activates a sthāyin that

he shares with all kings and all subjects, and he is soon

reliving their love less intensely but more richly and

serenely.

Objective correlatives are primarily representations in art

of the actual causes (laukika-kāraṇas) and the actual con-

sequences or manifestations (laukika-kāryas) of an emotion

in life. Representations of causes are called vibhāvas and repre-

sentations of manifestations are called anubhāvas. The actual

factors of an emotion in life are transformed thus into the

conventional associates of the same emotion in art. They have a purely aesthetic existence-they are not real or practical, but idealized; they are not personal or particular, but universalized; they have no conative drift and are objects of a detached untroubled contemplation that does not issue in action. When the sensuous objects of an emotion, thus universalized in art, impinge upon an emotional set latent in the reader's consciousness, the two coalesce and give rise to the state of utter aesthetic satisfaction known as rasa.

But what precisely is the nature of the relationship between the sensuous correlates and the emotion-the nature of the process by which they give rise to the emotion? Bharata (c. Ist century), whose Nātyaśāstra puts forward the earliest known exposition of rasa, had declared laconically that a conjunction of the correlates of emotion give rise to the rasa. How exactly it did so he had not explained-he left that to the commentators! The first of these, Lollaṭa, offered the naturalistic version: in art, correlates produce an emotion exactly as actual causes do in life; the sorrow that fills the Sundarakānda of the Rāmāyana is felt by the reader as a sorrow residing in a real Sītā. Śaṅkuka offered a logician's account: the relation of the object to the emotion is the same as that of the middle term (sādhana) to the major term (sādhya)-one, that is, of inference; the realization of rasa is an act of cognition, made 'relishable' by the charm of art. These explanations missed both the essential subjectivity of the rasa experience and the essential distinction between objects in art and objects in reality. Bhaṭṭanāyaka described the process of realizing rasa as comprising the universalization of experience and the elevation of the emotion (within the reader) to a state of ideal joy.

SUGGESTION: OBJECTIVE CORRELATIVE

39

This was a valuable advance, but it left the process of rasa-realization still obscure.

It was given to Abhinavagupta to achieve the breakthrough in this controversy about what Bharata's aphorism on rasa-realization meant. The emotion, he said, was already there within the reader—as the sthāyin. The correlates therefore do not produce or generate it but only render it manifest. Emotion emerges from its descriptively presented correlates exactly as the suggested meaning (vyañgyārtha) emerges from the stated meaning—by the operation of the function of suggestion (vyañjanā) which is inherent in language. Emotion is suggested meaning.

When the realization of rasa was thus explained in terms of dhvani (suggestion of meaning), the whole phenomenon swung into focus. It became clear that the emotion is the image's resonance—and not reference, nor inference. Reference is denotation—a given word standing simply, precisely and invariably for a given referent. Suggested emotion, on the other hand, is in the nature of connotational meaning—complex, rich, relatively imprecise, variable. Object-emotion associations for members of the same culture are, of course, bound to have a high common factor, but they can never have the precision and universality of denotational meanings. The object of an emotion thus is not a referential sign, but an evocative symbol. Nor is the realization of the emotion a process of inference. Given the object, the emotion does not necessarily follow. The object is not bound to the emotion by a cause-and-effect relation or by invariable concomitance. The object is only the suggestive associate of the emotion.

Dhvani thus explains rasa; in fact the two together (rasa-dhvani) are the supreme mode of poetic expression. Perhaps no other case exists of one major critical tenet lighting up

D

40

SUGGESTION: OBJECTIVE CORRELATIVE

another so much, of two independent critical traditions enriching each other so much and eventually coalescing. Its affirmation that all emotion in poetry is suggested and its version of how this is done are, I think, Sanskrit’s most valuable contribution to the theory of poetry.

From the principle that the main business of poetry is to present emotion by suggestion, Sanskrit theory was led naturally to the principle that every poem should have emotional unity. Ānandavardhana insists that a poem, however long, elaborate or diversified, is ruled by a single principal rasa. The Rāmāyaṇa, consisting of 24,000 verses, is governed by a principal emotion. So is the Mahābhārata, consisting of 100,000 verses. (That the presiding rasa of the Mahābhārata—an epic that tells of the greatest war of Indian legend—should be found to be śānta (serenity or quietude) and that of the Rāmāyaṇa—a poem which, if the Uttarakāṇḍa is not admitted to the Valmīki canon, ends with the restoration of Sītā to Rāma and of Rāma to the throne—would (I think) still be karuṇa (pathos) shows how the concept of rasa can uncover an undercurrent of feeling which the surface life of the poem belies.) All that happens in the poem—rhythm, language, imagery, action, character—must be contributory to the final cumulative emotional effect. Does this mean that the tyranny of the presiding emotion imposes on a poem a homogeneity wholly alien to it, inhibiting the amplitude and diversity of an art organism’s life? A. B. Keith, for instance, finds Sanskrit drama over-stylized and wooden and puts this down to the forcible orientation of everything in the play to a single central rasa.8 But it need not be, it indeed never is, just one emotion. The principal rasa, itself rich, diversified and complex, is fed by a multiplicity of minor incidental feelings (vyabhicārins), and also

accommodates other rasas which are subordinate to it in the

poem, being developed in this or that part of it and dropped

when done with. In the Śākuntala, love is the master motif,

but there are other motifs—sorrow, tenderness, serenity,

laughter. The principal emotion is thus one among many

and derives its special status only from the fact that it

permeates the poem, the poem indeed sums itself up in it—

and also from the fact that the other emotions, while freely

and fully orchestrated, are carefully harmonized with it by

assimilation, by balance or by tension. If this were not so,

if the poem were an essay in a single emotion, it would be

unendurable. A poem is essentially a system of emotions,

and the unity which the principal rasa establishes is a unity

in diversity.

A quick résumé like this cannot do justice to a critical

system which investigated in profuse detail the suggestion

of emotions through their objects and found this to be the

central business of poetry. But the summary may prove

handy when we go on now to enquire whether the system

which the Dhvanyāloka represents has any real points of

contact with, or important implications for, the tradition to

which Eliot's essay on Hamlet belongs.

The concept of nine or more sthāyins or permanent

emotions seems to be an oversimplification beside what we

have learnt (from psychology, psycho-analysis and art) about

our being's darkest and deepest region—the unconscious

and the instinctual—with its minute differentiation and in-

finite proliferation, and beside what the Symbolists have

taught us about the uniqueness of each fugitive tone and

shade of emotion. Yet psychologists and poets have in fact

often tended to talk in terms of emotional stereotypes. In

42 SUGGESTION: OBJECTIVE CORRELATIVE

the early 'twenties, William McDougall attempted an enumeration of the 'Primary Emotions' in the descending order of specificity and distinctness, and the sthāyins, all of them but one, form the upper half of his list.9 And a poet like Robin Skelton can talk of poetry arousing 'love, anger, enthusiasm, scorn, pity, hate and laughter'.10 To try to isolate thus the primary forms of our emotional life is not to deny its darkness and immeasurability.

As for the doctrine of the singular nuance or instant of emotion, it led, as we know, to a self-conscious precision of expression, but also, in certain cases, to failure of communication. The felt urge now for rebuilding the bridges, the new awareness of the social function and destiny of imaginative writing, the concept of imagination as race-experience—all these have taught us to regard what is shared as at least as important as what is incommunicably personal. No one today will claim for poetry that it is wholly an awakening of universal dispositions or wholly an expression of unique sensibility—the two run ever into each other. The exponents of Rasa-dhvani insist that presentation in art individualizes a universal emotion. The Dhvanyāloka explains how each time an emotion is presented again, the process of suggestion renders it new. The character of the emotion as it gets expressed is uniquely determined by the 'conjunction of correlatives'. Since no set of objects can be identical with another, an emotion, each time it is presented, necessarily offers a different refinement or facet of itself.

The first important concept of the Rasa-dhvani theory thus is the sthāyin, the subjective source of rasa. The second is that of the vibhāvas and anubhāvas—the objective correlatives of rasa. Bharata, who first formulated these, is

SUGGESTION: OBJECTIVE CORRELATIVE

43

separated from Eliot by some twenty centuries, and if Bharata’s vibhāvas and Eliot’s ‘set of objects’ are placed side by side, it may be difficult to recognize them as the same articles. In subsequent theory and, of course, in poetry, the vibhāvas were a great deal less naïve than in Bharata’s Nātyaśāstra, but we must nevertheless see them as they were first set forth by the Father of Sanskrit Poetics. For grief, Bharata’s correlatives are the death of the loved one, loss of wealth, captivity and exile. For love (‘love in union’ as distinct from ‘love in separation’), his correlatives are the season (spring), the loved one, the garden, garlands, unguents, ornaments. Compare these correlatives with MacLeish’s:

For all the history of grief

An empty doorway and a maple leaf

For love

The leaning grasses and two lights above the sea.

(‘Ars Poetica’)

Bharata’s correlatives are the immediate sensory equivalents of the emotion—they are direct, explicit, public, conventional. MacLeish’s are closer to symbols, more concentrated and potently suggestive, arbitrary and private and yet strangely eloquent, independent and free-standing, being neither naturally nor traditionally connected with the emotions they evoke. Of the symbol as such—risen from the unconscious, non-discursive, autonomous, productive of manifold meaning—Sanskrit criticism was largely innocent. (The alamkāra (trope) is, of course, a product of the imagination and, according to Ānandavardhana, is a servant to suggested emotion, and where based on metaphor, it comes pretty close to being an image, but it is at best a figure and at worst—as the name implies—an embellishment.) MacLeish’s leaning grasses are a far cry from Bharata’s unguents and

44

SUGGESTION: OBJECTIVE CORRELATIVE

garlands and even from Kālidāsa’s richly evocative jasmine

creeper and bee which, relatively sophisticated, are never-

theless firmly traditional. Yet the difference is only one of

degree, and all these objects, in varying degrees, answer to

Eliot’s description of the objective correlative: ‘. . . when the

external facts, which must terminate in sensory experience,

are given, the emotion is immediately evoked.’ Bharata’s

garlands are closer to the leaning grasses than to the garlands

at the florist’s.

Sanskrit criticism does not let us forget for a moment that

the poet’s garland and the florist’s are two qualitatively

different entities. How objects of emotion in art differ from

real objects has already been touched upon, but what is

the actual process that effects this transformation? Sanskrit

aesthetics has the same answer as Western aesthetics: uni-

versalization. Now this is an overworked term in aesthetics

and can mean either a great deal or very little. Clichés like

‘de-personalization’ and sub specie eternitatis and ‘the concrete

universal’ are not of much help. How in Eliot’s own case

private images became objective correlatives with a general

significance, his critics have not been able to explain. One

of them thinks Eliot hit upon potentially universal images by

chance;11 Eliot himself feels that ‘consciously concrete’ images,

if ‘clearly rendered’, become ‘unconsciously general’.12

This is not of much help either. A great deal of important

work has, of course, been done on the growth of facts into

symbols, not in the hands of an individual poet but by a

slow process of evolution. A version of this process which

seems to have special relevance to the theory that emotion is

suggested through its objects is offered by Wimsatt and

Beardsley in the essay I quoted from earlier. They point out

how ‘fictitious or poetic statement, where a large com-

ponent of suggestion (and hence metaphor) has usually appeared' creates 'out of a mere case of factual reason for

intense emotion a specified, figuratively fortified and permanent object of less intense but far richer emotion'.13 By

attaching emotion to object (horror to the murder of Duncan, or pathos to Shylock) poetry confers on both a permanence

which enables them to persist through changes in culture. The historical Cleopatra is a fact, a cause in reality of

Marcus Antoninus's infatuation. Shakespeare's Cleopatra is a symbol of love. The former is the actual cause of an

emotion, the latter a suggestive associate of it. This is how stories grow into recurrent myths—objective universalized

forms of expression for experiences that are conceived inwardly and uniquely.

Yet no account of universalization which regards it as a way of treating the material of poetry will ever be quite

satisfactory. Sanskrit theory defines universalization not as a process to which the poet subjects his material but as some-

thing which goes on inside the reader's mind. According to Abhinavagupta, universalization (sādhāraṇīkaraṇa) is brought

about by qualities of style, figurative expression and rhythm and (in the theatre) also by music, song and dance. These

obviously affect not the poet's material but the reader's or spectator's mind which then finds itself withdrawn from

reality into a world of formal beauty where the real, the personal and the particular seem to have no place and it is

easy for him to realize Cleopatra within himself not as a glamorous Ptolemy of the Ist century B.C. but as a woman

who loved and was loved, indeed, in a way, as all women who ever loved and were loved.

The enumeration of the traditional objective correlatives of given emotions in Sanskrit criticism would make one feel

46

SUGGESTION: OBJECTIVE CORRELATIVE

that the poet starts with an emotion to express and looks round for its external equivalents. This is, in fact, according to Graham Hough, the case against Eliot’s principle. ‘I wish’, says Graham Hough, ‘to point to . . . the suggestion that the whole natural world offers to the poet a collection of bric-à-brac from which he takes selections to represent emotional states . . . Plainly an eccentric view of the poet’s procedure . . . Gerard Manley Hopkins wrote “The Wreck of the Deutschland” because he was moved by the account of a shipwreck in which five nuns were drowned; he did not go round looking for a suitable disaster to match an emotion that he already had.’14 Raymond Williams agrees that ‘Mr Eliot’s statement of the matter implies an ordered process, in which the particular emotion is first understood, and an objective correlative subsequently found for it’. But (he observes) in another sense, objective correlatives ‘may serve as a precipitant to the artist, in that through their comprehension the artist is able to find a provisional pattern of experience . . . Finding the objective correlative may often be for the artist the final act of evaluation of the particular experience, which will not have been completely understood until its mode of expression has been found.’15 Whatever it be that happens at the poet’s end, Sanskrit theory is quite clear that in the reader’s consciousness the set of objects and the emotion arise almost, if not absolutely, simultaneously. The objective and the subjective are thus inseparable, co-extensive. The suggested meaning (the emotion) springs instantaneously from the stated meaning (the objects), the process being so rapid that its stages are imperceptible (asaṃlakṣyakrama).

The third important concept of the Rasa-dhvani theory

—that of a dominant emotion holding the whole poem

together—is not unique to Sanskrit. 'Unity of impression',

'emotional unity', 'total response' and several other formulations,

which are similar to, if not identical with, that of the

dominant rasa are met with in Western criticism. The oldest

and most prestigious of these is, of course, that of pity and

fear and of tragedy through them 'effecting its katharsis of

such emotions'. It is tempting to equate pity and fear with

the karuṇa and bhayānaka rasas, but what Aristotle meant by

emotions here is clearly different from what Ānandavardhana

meant by rasas. Pity and fear, as seen by Aristotle, are not

emotions presented in the play; they are emotions called up

in the reader or spectator by what the play presents—

emotions, that is, which are his reactions to the emotions

presented in the play. According to the rasa theory, on the

other hand, the emotion which the reader experiences is the

same as the emotion presented in the poem—a heightened

version certainly, but essentially the same emotion. The

emotion presented by the poet through its objective cor-

relates; the permanent emotion (sthāyin) awakened in the

reader; the rasa he finally experiences—all three relate to the

same emotion. It is therefore not like Aristotle's pity or fear,

a reaction or sequel to the experience. There is (for reasons

that need not be gone into here) no tragedy in classical

Sanskrit drama, but if it had accommodated tragedy, the

tragic emotion would have been not pity for the sufferer in

the play, but the suffering itself (śoka) developed into the

karuṇa rasa.

When Sanskrit theory speaks of a single emotion domin-

ating a dramatic or narrative poem, what is meant is

obviously the emotion presented as resident in the central

character, motivating the plot and informing the language

48

SUGGESTION: OBJECTIVE CORRELATIVE

and imagery of the poem. Eliot's analysis of the emotional

motif of Hamlet, like his definition of the objective correla-

tive, is entirely consonant with the Rasa-dhvani theory.

The essential emotion of the play, as he sees it, is the emotion

that dominates the Prince. This emotion, bībhatsa (disgust),

and not pity or fear, is the dominant emotion of the play.

The basic tragic emotion, karuṇā, is also at work, but it is not

the supreme principle of the play as it is of King Lear.

The emotional quality of a poem is often determined by

the emotional composition of its hero. Achilles and Jimmy

Porter are ‘hot, impatient, revengeful, impiger, iracundus, in-

exorabilis, acer, etc.’ (to quote Dryden's summing up of the

hero of the Iliad), and we have Homer's epic and Osborne's

play, disparate in all else but both ruled by raudra (anger).

(In the Iliad, raudra is reinforced with vīra, the heroic.)

Aeneas is (so is Yudhishthira) ‘patient, considerate, careful of

his people, and merciful to his enemies; ever submissive to

the will of heaven, quo fata trahunt, retrahuntque, sequamur’

(Dryden's summing up again), and we have the Aeneid and

the Mahābhārata, obeying an undercurrent of śānta (serenity).

But is it always as simple as that? Cannot a poem be

telling at times of an experience that is not strictly emotional

or is a mixture of affective and cognitive elements or is so

complex and individual that it cannot be classified at all?

It is, of course, idle to pretend that every poem, every line of

a poem, is touched with one or the other of the primary

emotions. But the point surely is that a poem which does not

want to disintegrate into several poems or a play which

does not want to ignore the fact of a theatre audience has

to orient itself firmly to a final cumulative emotional im-

pression. This may or may not be one of the traditional

formulations but must essentially be felt as a distinct and

SUGGESTION: OBJECTIVE CORRELATIVE 49

integral experience. Sanskrit theory does not agree that the

suggested meaning (whether it be emotion or any other form)

can be obscure or indeterminate (anirdeśya).

One of the chief concerns of the Rasa-dhvani theory has

been the study of how the other rasas in a poem are organ-

ized round its dominant rasa. Some of the richest effects in

poetry are wrought from the juxtaposition or alternation,

the opposition or union, of different rasas. The meaning of

Samson Agonistes, for instance, lies less in theology or auto-

biography than in the contrapuntal employment of vīra

(the heroic) and karuna (the pathetic). The raison d'être of the

plot is to conduct Samson from sorrow (śoka, the base of

karuna) to a reborn elation and ardour (utsāha, the base of

vīra). From the plane of retrospective contemplation where

it dwells in the poem, for it is alien to the realities of Samson's

present, vīra acts upon the poem's chief emotion, karuna

(the quality of the present), penetrating and assimilating it—

though the mood at the actual close of the poem is śānta

(calm). To describe the poem in these terms is not to de-

humanize it but to unfold the design beneath the play of

personalities and events. The point, similarly, of St Peter's

speech in Lycidas is not that it is the customary dose of

satire in a pastoral elegy nor that it is really Milton de-

nouncing the contemporary clergy. It is undoubtedly both

these, but its meaning, just where it is placed, lies in its note

of raudra (indignation)—or is it bībhatsa (repugnance)?—

which, by tension, reinforces the elegiac emotion (karuna) to

whose tokens of graceful memorial tenderness we revert

when the dread voice is past.

It seems presumptuous to take the floor in the Satan debate

and then only to suggest that it need perhaps never have

been started. Given that emotion supplies the conscious or

50 SUGGESTION: OBJECTIVE CORRELATIVE

unconscious design of poetry and that a whole poem is governed by a single emotion and different parts by different emotions, 'Satan, hero or fool?' is perhaps too simple a form in which to propound the question. The ruling emotion of a poem which tells how disobedience brought death into the world and all our woe is, of course, karuṇa. But the opening portions of the poem present a different woe—the Apostate Angel's deep despair—from which soon enough comes the blaze of high defiance. The dominant emotion of the first two books is the heroic (vīra), and Satan is, while it lasts, its chief objective equivalent. When Milton passes on to a different emotional key, Satan's heroic proportions inevitably shrink. Since consistency is to be defined not as continuing sameness in the character but as accord of character with emotion, the Archfiend 'rotting away' is a process of no interest to the saḥrdaya, the reader who seeks to be en rapport with the emotion.

A major problem of contemporary poetry and poetics is that of how a long poem in the strict Symbolist manner, composed exclusively of non-propositional imagery, can exist as a whole poem and not as a mere string of symbols. But the solution was presented when the problem itself was first presented in The Waste Land. The Waste Land is a disposition of symbols. It is also a long poem with as much unity as Aristotle or Ānandavardhana could have wished. Its inconsequential centrifugal images are held together by an underlying emotional unity—a single dominant rasa working as an undercurrent. It is impossible to identify the specific rasa of The Waste Land as one of the 'permanent emotions'. It is in fact impossible to find a name for it—Eliot's readers called it disillusionment, and Eliot thought it could be their own 'illusion of being disillusioned'. It is all

the same a real enough emotion and constitutes the total meaning of the heterogeneous images that make up the poem. In the introduction to St John Perse’s Anabase, Eliot himself confirms that ‘the reader has to allow the images to fall into his memory succesively without questioning the reasonableness of each at the moment, so that at the end, a total effect is produced’. The total effect is an emotional effect; as Richards says, speaking of some of Eliot’s poems, ‘ . . . the items are united by the accord, contrast, and interaction of their emotional effects, not by an intellectual scheme that analysis must work out. The value lies in the unified response which this interaction creates in the right reader.’16 In other words, emotion is structure.

4

The Lamp and the Jar: Stated and Suggested Meaning

One of I. A. Richards's more significant latter-day insights —or recantations?—is the notion that prose (scientific language, that is) is a questionable means of studying and describing the not-prose and that metaphor can be seriously used 'not as a literary grace, but as a technique of reflection and an operation of research'.1 Curiously enough, the thinking and talking about poetry that went on in Sanskrit criticism did in fact turn upon crucial metaphors. The Anvitābhidhāna theory of meaning used the metaphor of the arrow; the theory of Rasa used, among other things, the concoction of pepper, candy and camphor as a tool metaphor. But the 'founding metaphor',2 to apply Richards's phrase, of the central poetic in Sanskrit—the Dhvani Poetic —is the lamp and the jar. The metaphor identifies the two factors of suggestion as the suggestor and the suggested. On the one hand, there is stated meaning, which is the suggestor; and on the other, there is suggested meaning. The lamp is stated meaning (vācyārtha); the jar is suggested meaning (vyañgyārtha) which it illuminates and reveals. The metaphor was used by Sanskrit critics, although probably not quite in the manner Richards has in mind, as an instrument for analysing and describing the relationship and comparative status of stated and suggested meaning. This, I think, is a

52

unique enquiry; a full-length study of poetic meaning as

dual and hierarchical has not been, as far as I know, under-

taken elsewhere. The explanation for this is the reign of

certain other metaphors ! ‘ . . . A Swedish drill, in which

nothing is being lifted, transported, or set down, though the

muscles tense, knot, and relax as if it were’ (Donald Davie,

explaining Susanne Langer’s theory);3 the ‘bit of nice meat

for the house-dog’ that ‘the imaginary burglar is always

provided with’ (T. S. Eliot)4—stated meaning, as these

metaphors used for describing it show, is often looked upon

as an illusion or as a diversion, and the attitude has inhibited

any serious study of the stated and the suggested as the two

dimensions of meaning. Another obstacle has been termin-

ology. The word ‘meaning’ is sometimes used to denote

what is stated—as in F. W. Bateson’s summing up of the

Suggestion Theory: ‘ . . . poetry is not concerned with what

words mean but with what they suggest. A good poem

apparently should mean very little while suggesting a great

deal.’5 ‘Meaning’ at other times is made to denote what is

suggested, as when John Bayley speaks of searching for

Dylan Thomas’s meaning and equates it with ‘a concealed

significance’.6 Worst of all, ‘meaning’ can be used to denote

stated meaning one moment and suggested meaning the

next, as when T. S. Eliot, within the space of 28 lines, first

speaks of meaning as explicit prose discourse and then refers

to ‘a meaning which reveals itself gradually’ in Shakespeare’s

plays.7

Such inadequacy of terminology can frustrate any com-

parative study of stated and suggested meaning. But a more

serious obstacle is the view that the dichotomy is unreal and

that meaning is one and indivisible. To return to John

Bayley on Dylan Thomas, he deprecates conscious ex-

54

THE LAMP AND THE JAR:

ploration of suggested meaning and says: ‘There is no gap—

no intellectually sensible gap thatis—between our grasping of

the words and our deduction of what they are supposed to

stand for.’8 To be sure, according to Sanskrit theory, there

is in practice no gap between the grasping of the statement

and the grasping of the emotive meaning suggested; that is

to say, the mind leaps it so instantly that the gap is not

sensible (asaṃlaksya) and the two meanings—the stated and

the suggested—are perceived almost simultaneously. Almost

but not quite—because essentially the apprehension of sug-

gested meaning succeeds (however imperceptible the stages

of succession) the apprehension of stated meaning. In per-

ception, the two meanings are almost one; in description,

however, they can very properly be treated as distinct

elements. We have after all (to use Pater’s phrase) an

achieved distinction here, and there is nothing to be gained

by obliterating it. On the contrary, viewing statement and

suggestion as separate entities and analysing their mutual re-

lation might help to light up much that is obscure and con-

fusing in the emergence of poetic meaning.

But we might be deepening the confusion instead if we

do not begin by defining the term ‘statement’. ‘Statement’,

as used in this chapter, means literal or stated meaning

treated as one component of a suggestive poem, the other

component being suggested meaning. In a later chapter I

will be using ‘statement’ to mean whole poems (or portions

of them) that are non-suggestive. As an explanation, if not as

an excuse, for this verbal looseness, I can plead that ‘state-

ment’, like ‘meaning’, is an instance of inadequacy of

critical terminology. I can also plead that I am not alone

here. Tillyard, for instance, in the very act of defining what

makes ‘The Deserted Village’ and ‘The Echoing Green’

STATED AND SUGGESTED MEANING

55

specimens of two opposed kinds of poetry describes 'The Deserted Village' as statement poetry and speaks of the statement in 'The Echoing Green' ! Sanskrit has separate terms for the two uses of the word 'statement'. The first 'statement' (i.e. discursive poetry) is citrakāvya; the second (i.e. stated meaning) is vācyārtha. In this chapter I use the word in the latter sense—the sense which the lamp vis-à-vis the jar denotes.

The use of the lamp-jar metaphor as an instrument of reflection in ninth-century Sanskrit criticism yielded several insights into the behaviour of the stated and suggested components of meaning when the suggested component is emotion and the stated component is its objective correlatives. Into their behaviour, that is, as observed in contemporary or earlier Sanskrit poetry, so that it can't be claimed for these findings that they have universal applicability. But some of them might prove useful in re-examining, in present-day terms, the relationship of stated and suggested meaning; and in attempting this we will not be quite so much describing these findings as using them as points of departure for an independent study.

In the first place, the jar was already there, the lamp only revealed it. The lamp and the jar are a conventional formula in Indian philosophy for stating the theory of manifestation. Applied to poetry, the theory sees the process of suggestion as the revealing of what is already there rather than as the presenting of anything new. 'What is already there'—one is reminded of the remark in Empson's Seven Types: ' . . . Being an essentially suggestive act it (poetry) can only take effect if the impulses (and to some extent the experiences) are already there to be called forth; that the process of getting to understand a poet is precisely that of constructing

E

his poems in one's own mind.'10 It is clear that Empson is thinking here of reader participation as a factor of suggestion, and this, of course, is a different thing from the ontological recognition (embodied in the lamp-jar formula) of the nature of suggestion as identical with the nature of manifestation. All the same, the principle of reader participation can be looked at in terms of the lamp and the jar—the stated meaning, when the reader makes contact with it, acts upon an existing capacity within him for response to symbolized emotion, and the suggested meaning is rendered manifest in his consciousness. Sanskrit theory, as we saw in the last chapter, assumes that when the sensuous correlates of the emotion activate one of the emotional sets latent in the reader, the supreme aesthetic response is evoked. It also assumes that only the reader whom nature or nurture has equipped with a special sensibility can respond in this way. As Lascelles Abercrombie says, ‘ . . . the author must rely on his readers' ability to respond to what his language can only suggest . . . It is the sense of language, proved by the enjoyer of literary art.'11 This type of reader goes to work actively on the poem—on this showing, being illuminated is not a passive rôle for the jar but an energetic act of response.

The second observed feature of the lamp-jar relationship is that the lamp continues to shine even after the jar has been illuminated. To class statement as stuff which gets used up in the generation of suggested meaning, as something that disappears from, or can be dismissed from, our thoughts when it has served its purpose is to misread the process of suggestion. Even when the suggested meaning has been apprehended, the stated meaning is intact and continues to be

active. Sanskrit critics thus saw statement as non-expendable,

except where metaphor is present and supersedes, as it must

(according to them), the stated meaning. (We will return

in the next chapter to the action of metaphor on statement.)

Where the stated meaning consists in the objective correla-

tives of the emotion, these, far from being superseded, con-

tinue to exist—in fact, they suggest the emotion by them-

selves developing rapidly into it. The granite shore, the

white sails and the cry of quail are not banished from our

consciousness when the mood that Ash Wednesday evokes

through them has been realized—they persist and are integral

with the mood.

But the mood can fail to get realized if its correlates are

inadequately or ineptly presented. When T. S. Eliot speaks

of lack of ‘explicit reference of emotion to object’,12 or lack

of ‘adequacy of the external to the emotion’,13 he is in fact

referring to inadequacy of statement; and when he calls

Hamlet a failure, he means that it is a failure of statement.

And this brings us to the third observation: the lamp not only

reveals the jar but reveals itself. It could not establish the

identity of the jar if in the very act of doing so it didn’t

establish its own, asserting its own solidity, presenting itself

as a bright clear object. Eliot too—and here is a remarkable

coincident insight—defines the relation between stated and

suggested meaning in terms of a light: ‘. . . suggestiveness is

the aura around a bright clear centre.’14 In order to have

the aura (the suggested meaning), you must have a bright

clear centre (the stated meaning). In order to illuminate

the jar the lamp must also illuminate itself.

Sanskrit theory makes the point that, strictly speaking,

the suggested meaning is apprehended, despite the apparent

simultaneity, after the stated meaning and indeed through

58

THE LAMP AND THE JAR:

it and is governed by it. Language, we may add, cannot

suggest unless it has first stated, and then what it stated is the

base which supports what it suggests. Sanskrit criticism calls

this vācyārthāpekṣā (dependence on stated meaning). As

F. W. Bateson points out, if you don’t know where Troy

is and what happened there, you can’t respond to the sug-

gested meaning in Yeats’s line, ‘Troy passed away in one

high funeral gleam’.15 This, of course, is but common

sense—the least we owe the poet is to begin by reading what

he wrote and understanding his plain meaning. If we neglect

to do this, we are in danger of erring as egregiously as Edith

Sitwell did when Dylan Thomas’s ‘Atlas-eater with a jaw

for news’, who ‘bit out the mandrake with tomorrow’s

scream’ from the gentleman’s fork, was taken by her as

referring to the modern craze for speed, horror and sensa-

tion. As John Bayley explains, she ignored the verb ‘bit out’;

no wonder Thomas complained, ‘She doesn’t take the

literal meaning’.16 When statement is read perfunctorily (as

in this case) or is tenuous and inchoate (as in some Romantic

poetry) or tortured and unintelligible (as in some Symbolist

and Post-Symbolist poetry), suggestion is aborted.

It is, of course, true that if inadequacy of statement can

abort suggestion, so can perfect adequacy of statement—

in a different way. This is a very necessary qualification

to attach to the vācyārthāpekṣā concept. Where the stated

meaning is self-contained and sufficient, it doesn’t compel

reference to anything beyond itself so that any further

meaning found is more a bonus added than a need met.

To return to F. W. Bateson, he finds in Herrick’s ‘A Sweet

Disorder in the Dress’ a covert plea for pagan morals and

manners. He arrives at this interpretation by taking the

human adjectives applied by the poet to features of dress and

referring them back to the wearer.17 It is an attractive interpretation, and if its authority seems to be less than adequate,

it is because the poem can be read quite satisfyingly as no more than an elegant compliment to a girl's charm which

only gains by any carelessness in dress; there is nothing in the poem to oblige us to get beyond this meaning which is by

itself sufficient and entire. Compare Herrick's poem with Roethke's 'I Knew a Woman':

Love likes a gander, and adores a goose:

Her full lips pursed, the errant note to seize;

She played it quick, she played it light and loose;

My eyes, they dazzled at her flowing knees;

Her several parts could keep a pure repose,

Or one hip quiver with a mobile nose

(She moved in circles, and those circles moved).

Unlike so much poetry that uses deformation of syntax as an instrument of suggestion, these lines are syntactically

regular, but all their other features—the logical discontinuity, the discrete images, the incongruity of certain collocations—

serve to make the statement incapable of standing on its own legs; the suggested meaning thus becomes an imperative

need. Where, on the contrary, the statement is independent and self-supporting, it always usurps the dominant rôle, and

we have (for a different reason from what Sanskrit theory had in view when it coined the phrase) 'the poetry of sub-

ordinate suggestion'.

Whether the deriving of suggested meaning from the stated is not after all a form of logical inference was the

subject of a lengthy debate which the Dhvani theorists had to fight and win before suggestion was accepted as a sub-

stantive mode. And it isn't as though these schoolmen went

60

THE LAMP AND THE JAR:

on sharpening their teeth on each other over an exquisitely

bookish point. The distinction between deduction and sug-

gestion is not always as obvious as one would think. We

are all close readers today and tend to be cerebral, so that in

expounding the unstated meaning of a poem we are not

above cheating—we can, that is to say, catch ourselves

offering a deduction wrapped and labelled as a suggestion.

Or we can find ourselves syllogizing unabashedly, as Pound

did, commenting on a Li Po piece translated by himself:

The jewelled steps are already quite white with dew,

It is so late that the dew soaks my gauze stockings . . .

Pound’s gloss is: ‘Jewel stairs, therefore a palace . . . Gauze

stockings, therefore a court lady . . .’18 But no cast-iron

‘because—therefore’ can make us move inexorably from

‘jewelled steps’ to ‘a palace’—jewelled steps could equally

legitimately suggest street lights on a steep rise, a starry sky

or a bedecked bosom. This brings us to the fourth finding:

the lamp and the jar are not necessarily found together. Inference

is based on the invariable concomitance of the middle term

and the major term; stated and suggested meaning have no

such relationship. Yet another statement possible about the

lamp and the jar is that although the jar gets lighted up after

the lamp has started giving off light, the interval is too

minute to be noticeable. A conclusion follows a premise by

perceptible succession marked by mediacy; on the other

hand, the illuminator-illuminable relation which governs

stated and suggested meaning is based on immediate and

imperceptible succession. The work of our best close readers

shows that they arrive at the further meaning by association

and not by inference. From Nash’s line ‘Brightness falls

from the air’, Empson derives the meaning that the earth is

brighter than the sky when the clouds are dark and there is

a threat of thunder.19 This, although ingenious, is not a logi-

cal conclusion, for Empson derives several other possible

meanings from the same line. From the words ‘cannot

scare’ in Frost’s line, ‘They cannot scare me with their empty

spaces’, Cleanth Brooks and R. P. Warren obtain the meaning

that the speaker is a mature man.20 This is not inferring the

state of being a child from the state of being scared; it is

just selecting one of the connotational meanings of ‘scared’.

Ninth-to-eleventh-century Sanskrit criticism produced a

considerable body of theory not only about the relation of

stated to suggested meaning but also about their comparative

status. Its main assumption, reduced to very simple terms,

is that stated meaning should be subordinate and suggested

meaning supreme: supreme by virtue of its aesthetic worth

and also by virtue of its exclusive importance (which implies,

I suppose, that stated meaning should have no importance

whatever except as suggestor). Suggestive poetry strictly so

called is where the suggested meaning is paramount; where

it is not paramount, we have merely gunībhūtavyaṅgya, ‘the

poetry of subordinate suggestion’. The principle, stated in

such general terms, seems unexceptionable, but when its ex-

ponents proceed to apply it to a given passage, one finds

that the grounds on which they determine the status of the

suggested meaning with reference to its beauty and im-

portance tend to be arbitrary, or at any rate unclear. Never-

theless it might be useful, quite independently of Sanskrit

criticism (although we will probably find ourselves return-

ing to it often), to try to devise guide lines for determining

the status of suggested in relation to stated meaning.

Suggested meaning reigns assuredly supreme when it

consists in an emotion or mood as distinct from an idea.

62

THE LAMP AND THE JAR:

Now this is, I am afraid, a coarse distinction worthy of the

young Yeats who spoke of symbols that 'evoke emotion

alone' and intellectual ones that 'evoke ideas alone'.21

As Cleanth Brooks and R. P. Warren point out, ' . . . a mood

implies an idea, as an idea implies a mood'.22 Yet it is perhaps

a distinction that embodies a real enough difference. The

point is that in a poem the mood created has necessarily

more authority than the sensuous details through which it is

created. The emotion-meaning suggested is essentially the

dominant principle in relation to the stated meaning,

because the objects presented descriptively are the correlates

of the emotion and cannot possibly outshine it without re-

pudiating their own nature. But when it is an idea, and not

an emotion, that is being evoked, the statement, I think, can

acquire an autonomous interest and dwarf what is suggested.

This can easily happen when an image develops so much

intensity and vigour that the idea of which it is the concretion

is quite overshadowed. This is the price Auden is sometimes

found paying for the nightmarish brightness, the cinemato-

graphic vividness, of his images:

For the wicked card is dealt, and

The sinister tall-hatted botanist stoops at the spring

With his insignificant phial and looses

The plague on the ignorant town.

('Epilogue' in Look, Stranger)

Sanskrit critics would call this 'the poetry of subordinate

suggestion'. Poe would have called it 'the poetry of the

fancy', adding: 'Here the upper current is often exceedingly

brilliant and beautiful; but then men feel that this upper

current is all.' If what Auden suggests in these lines is the

rôle of science as destroyer of civilization, the Hitchcockian

STATED AND SUGGESTED MEANING 63

effect succeeds only in focussing attention on itself and the

idea suggested is cast into the shade. In New Year Letter,

Auden arrives at the idea that there is a fearful price to be

paid if we try to prolong the momentary mystic release from

Becoming into Being:

He hears behind his back the wicket

Padlock itself, from the dark thicket

The chuckle with no healthy cause,

And helpless, sees the crooked claws

Emerging into view and groping

For handholds on the low round coping,

As Horror clambers from the well.

The picture is, at a lowbrow level if you like, so arresting

that what is suggested can’t hope to compete with what is

said. Earlier in the same poem, Auden offers a sort of micro-

film projecting the universal spiritual guilt of our time:

There lies the body half-undressed,

We all had reason to detest,

And all are suspects and involved

Until the mystery is solved

And under lock and key the cause

That makes a nonsense of our laws.

O Who is trying to shield Whom?

Who left a hairpin in the room?

Who was the distant figure seen

Behaving oddly on the green?

Why did the watchdog never bark?

Why did the footsteps leave no mark?

Where were the servants at that hour?

How did a snake get in the tower?

The situation, wrapped in evil and enigma, is elaborated

with some gusto; the features, which have no individual

64

THE LAMP AND THE JAR:

allegorical reference, add up to a total effect which, although

merely Agatha-Christian, is of such compelling vividness

that the concept suggested is quite eclipsed.

Like Auden's glowing images, some of Ted Hughes's

fauna--the tom cat, the too dead pig, the too slow bull--

are unforgettable studies that we start caring for their

own sake, and they seem to be more important than any

human implication that can be assigned to them. Yeats's

haunting image of the rough beast slouching in the sand

amid the desert birds clearly matters more to us than anything

it could be suggesting, and so does Shelley's Magus Zoroaster

who 'met his own image walking in the garden'. Eliot's

another one walking beside you

Gliding wrapped in a brown mantle, hooded

I do not know whether a man or a woman...

(The Waste Land)

is yet another image that states too vividly. I am not sure

this is always to be treated as a flaw. 'The symbol,' says

Tindall, 'is more important than what it suggests...'23

The self-validating symbol risen from the unconscious was

alien to the daylight beauty and order of classical Sanskrit

poetry on which the ninth-century theory of poetry was

reared; the presentation of identifiable emotion through its

traditionally associated objects was more in its line, and it

naturally saw as poetically vital not the symbol that tries

to present an experience, or maybe outshines it in so trying,

but the emotion that is evoked by its object.

Apart from the dazzling statement that steals the show,

there is another condition that can lower the status of sug-

gested meaning. This is a condition frequently encountered

in poetry given to explicitness: what has been suggested

STATED AND SUGGESTED MEANING

65

is also stated, and inevitably the spelling out reduces the

potency of the utterance. Take the following lines from

Gascoyne’s ‘A Wartime Dawn’:

Draw now with prickling hand the curtains back;

Unpin the blackout-cloth; let in

Grim crack-of-dawn’s first glimmer through the glass.

All’s yet half-sunk in Yesterday’s stale death,

Obscurely still beneath a moist-tinged blank

Sky like the inside of a deaf mute’s mouth . . .

Nearest within the window’s sight, ash-pale

Against a cinder-coloured wall, the white

Pearblossom hovers like a stare; rain-wet

The further housetops weakly shine; and there,

Beyond, hangs flaccidly a lone barrage-balloon.

An incommunicable desolation weighs

Like depths of stagnant water on this break of day . . .

The passage first calls up a mood through evocative land-

scape, but the last two lines, which are a statement of the

mood, succeed in depressing the status of the suggested

meaning. To realize how disastrously they succeed, one has

only to compare Gascoyne’s description with Eliot’s pre-

sentation of the same hour in ‘Preludes’ where a like mood

of desolation is evoked through descriptive detail but left

unnamed.

By calling the last two lines from Gascoyne a statement,

I am afraid I have committed the same offence that I

charged Dr Tillyard with at the beginning of this chapter—

that of simultaneously using the term ‘statement’ in its two

different senses. In this chapter we have been examining

suggested meaning in relation to statement-as-suggestor

(which is what the first eleven lines from Gascoyne are)

66

THE LAMP AND THE JAR

and not in relation to explicit statement (which is what the last two lines are). Yet we can't help concerning ourselves with the latter—because the point made here is that when a poem has presented a mood by suggestion, explication deprives the suggested meaning of its dominant rôle in relation to the stated meaning. That a symbol suggests what by its nature can never be stated is a familiar law of modern Western aesthetics; ninth-century Sanskrit critics were able to anticipate it. That they could do so without having known the principle of symbol is, I think, a remarkable achievement of insight.

5

Suggestion through Metaphor

An interesting finding of the imagery studies of recent years

is the distinction between the image which makes a point,

directs your attention to the specific relation between the

two terms of the metaphor, and the image which works by

suggestion, by the radiation of a quantity of multiple im-

precise meaning—between the stiff twin compasses and the

sick rose. The two are treated as mutually exclusive kinds,

labelled, and even identified with periods. ‘ . . . The meta-

physicals and the modernists stand opposed to the neo-

classic and Romantic poets on the issue of metaphor,’ says

Cleanth Brooks.1 ‘The making of a point, the outlining of a

picture, as opposed to that suggestiveness of image which is

the aim of the contemporary poet, as it was of the Romantics

and often of the Elizabethan dramatists, is the keynote of

Augustan imagery,’ says Cecil Day Lewis.2 So where are we?

Is Augustan imagery (which ‘makes a point’, according to

Day Lewis) akin to Metaphysical imagery (which also does,

according to Rosemond Tuve) or to Romantic imagery

(which, along with Augustan, eschews the play of the intel-

lect, according to Cleanth Brooks)? Is the modernist image

sprung from the Romantic image (as Read and Kermode

show) or from the Metaphysical image (as several others

show)? And do Elizabethan metaphors suggest, or do they

make a point?

67

68

SUGGESTION THROUGH METAPHOR

Obviously, dividing imagery into two kinds and linking them with periods scarcely helps. The value of Allen Tate's theory of 'tension' lies in its demonstration of the truth that it is only in bad poetry that imagery is exclusively 'extensive' (denotative) or exclusively 'intensive' (connotative or suggestive).3 Specificity either of statement or of metaphor, on the one hand, and richness of suggestion, on the other, are the two poles in all poetry, and every poet starts at either end and works towards the other, and if he pushes far enough but not too far, achieves 'tension'. The only difference between the Metaphysical and the Romantic is that they start at different ends of the scale, the extensive and the intensive—but their goal is the same: the point of tension. Metaphor and suggestion thus, far from being two categories of imagery, are two co-present functions at work within an image.

Co-present, but always distinct. The operation of metaphor proper—consisting ostensibly in the establishment of the logical relationship of the primary and secondary referents and in the vehicle's illumination of the tenor—is distinct from the accompanying emergence of imprecise expansible meaning. Indeed the latter can often be independent of the metaphor and vastly more important. 'More usually,' says I. A. Richards, 'the elucidation is a mere pretence . . . There are few metaphors whose effect, if carefully examined, can be traced to the logical relations involved. Metaphor is a semi-surreptitious method by which a greater variety of elements can be wrought into the fabric of the experience.'4 Others too have noticed the same curious phenomenon. 'In understanding imaginative metaphor we are required to consider not how B (vehicle) explains A (tenor), but what meanings are generated when A and B are confronted or

seen each in the light of the other . . .', says Wimsatt, explaining Martin Foss's theory of metaphor; and he quotes Foss as saying about the proverb 'Among blind men the one-eyed is king': 'The true significance of the proverb goes far beyond the blind, the one-eyed, and the king: it points to a wisdom in regard to which the terms of comparison are only unimportant cases of reference.'5 (An interesting parallel case of suggested meaning being independent of the image is Archibald MacLeish's 'coupled images'. When two unrelated images (such as a bracelet of bright hair and a bone) are juxtaposed, ' . . . a meaning appears which is neither the meaning of one image nor the meaning of the other nor even the sum of both but a consequence of both—a consequence of both in their conjunction, in their relation to each other.'6 This, of course, is different from what happens in a metaphor. In a metaphor the two terms are linked.) To return to Richards, explaining his notion that 'metaphor supplies an excuse by which what is needed may be smuggled in'—an application to metaphor of his theory that in poetry referential elements are only a means to the evocation of attitudes and are in themselves relatively unimportant—Richards speaks of a mysterious effect in art: an essential or important thing is said best if it is said not overtly and evidently but apparently inadvertently, as a by-product or an accidental concomitant.7 This is an important principle behind suggestion, but it throws no light on why metaphor often accompanies or precedes suggestion and whether and how metaphor serves suggestion. How is it that the complex meanings which surface around or in the wake of the metaphor are more important than the metaphor and apparently independent of it and indeed turn out to be the poet's real concern?

70

SUGGESTION THROUGH METAPHOR

Part of the answer seems available in Sanskrit poetics,

which distinguishes three chief meaning-functions of lan-

guage—statement (abhidhā), metaphor (lakṣaṇā), and sug-

gestion (vyañjanā). Metaphor, according to Sanskrit poetics

and philosophy, occurs when the primary referent of the

word is found incompatible with the intended sense in the

context, and to remove the incongruity (mukhyārthabādha),

meaning is transferred to a secondary referent connected with

the primary one by similarity or some other relationship.

The primary referent must be found incompatible with the

context; the secondary referent must be related to the

primary one; the transfer of meaning must have the sanction

either of usage (rūḍhi) or of the specific purpose or use

(prayojana) of the metaphor—these are the three basic con-

ditions of metaphor. In Donne’s line from ‘The Autumnal’

I shall ebb out with them, who home-ward go

the primary referent of ‘ebb’ is (for the tide) ‘to flow back’

and is inapplicable to a human being; so a secondary referent

(let us say ‘ageing’—when it is poetry you are dealing with,

secondary referents are at best clumsy substitutes) has to be

found. This transfer of meaning is ‘sanctioned’ or motivated

by the poet’s intention to suggest the serenity of love in late

life.* On the other hand, the historian’s cliché in ‘The

fortunes of the Eighth Army ebbed rapidly’ is sanctioned

by usage. Naturally, the incompatibility (anupapatti) of the

primary reference is more pronounced in an intentional or

created metaphor (prayojanavatī-lakṣaṇā)—created, as mostly

happens, by a poet—than in a conventional metaphor em-

*Wellek and Warren in their Theory of Literature (p. 196) refer to

Wilhelm Wundt’s notion that ‘the calculated, willed intention of its user to

create an emotive effect’ is the criterion of true or poetic metaphor.

SUGGESTION THROUGH METAPHOR

71

bedded in usage (nirüdha-lakṣaṇā) where the primary reference is faded. Why does a poet create a metaphor? The

raison d'être or purpose or use of the metaphor—that is to

say, what caused it at the poet's end; what results from it, at

the reader's—is revealed or apprehended by the process of

suggestion. Thus Donne's metaphor, having established the

relation between ebbing and ageing, goes on to suggest the

unstrenuous serenity of love in late life, almost indistinguish-

able from easy dissolution; and working associatively it links

itself with the earlier metaphor:

may still

My love descend and journey down the hill.

Of the meaning of a passage containing a metaphor, just as

much as is needed for removing the incompatibility of the

primary reference is the metaphorical or secondary meaning

—the rest is suggested meaning.

Let me pour forth

My tears before thy face, whil'st I stay here,

For thy face coins them, and thy stamp they bear,

And by this Mintage they are something worth . . .

('A Valediction: Of Mourning')

In the last two lines as they stand, the numismatic terms, on

the one hand, and the lover's tears and the mistress's face,

on the other, are in conflict with each other. When we

render the meaning of the lines as 'Because my tears will

reflect your face and thus acquire value', we are offering the

metaphorical meaning which resolves the disharmony. But

Donne's lines have a large residue of suggested meaning—

in beauty and worth and in her power over him, the mistress

is as a queen; there is also the tragic paradox of the currency

(what can be more permanent than something stamped on

F

72

SUGGESTION THROUGH METAPHOR

metal?) lasting only while she is physically before him.

There are, similarly, the legal metaphors in Shakespeare’s

lines:

The Charter of thy worth gives thee releasing;

My bonds in thee are all determinate.

(Sonnet 87)

When the primary references of the legal terms are superseded

and the secondary or metaphorical meaning of the lines given,

their meaning is far from exhausted. There is an envelope of

suggested meaning left: the irony of running an intimate

emotion into absolute impersonal moulds; the seal of heart-

breaking irrevocability which the definitive language of

law sets on the parting; and maybe much else. This, rather

than the metaphoric point, must be what made Shakespeare

employ legal imagery here. And here, as in Donne’s lines,

the suggested meaning is more or less independent of the

metaphoric point.

The nuances and insights of this relatively unspecific

suggested meaning are beyond the language of metaphor.

Poetic metaphor thus leads to and fulfils itself in suggestion,

but the two are distinct functions. When the apparent in-

compatibility of the vehicle with the context is corrected

by linking it to the tenor, the function of metaphor has

exhausted itself. The comparatively imprecise and fluid mean-

ing through which the purpose or use of the metaphor

manifests itself is extraneous to metaphor. There are many

more reasons why suggestion cannot be treated as part of

metaphor. Metaphor is based on the primary meaning, it

is the primary meaning transferred, extended; suggestion, as

in music, can occur in the absence of the primary meaning.

Metaphor supersedes the primary meaning; suggestion exists

alongside of it. Metaphor succeeds the primary meaning; the primary meaning and the suggested meaning arise almost (almost, though not quite, as we saw in the last chapter) simultaneously, as when a poem suggests an emotion by stating its objective correlatives. Metaphor specifies an idea, a logical relation; suggestion is imprecise, indeterminate, accessible through interpretation, and dependent on such variables as the writer, the reader, the context . . . These four or five sentences read rather like an undergraduate's notes and make the whole thing look too easy, but no summary can do justice to the formidable body of theory thrown up by that long shrill debate in Sanskrit poetics: Is suggestion distinct from metaphor?

The notion of 'Semantic Synthesis' propounded by F. W. Bateson is an attempt (analogous to Mukulabhaṭṭa's in Sanskrit which did not find acceptance) to subsume suggestion under metaphor.8 'Incantation' or 'verbal magic' is to Bateson the collocation and reconciliation of apparently unconnected, even contrasted or conflicting, units of meaning. Chaucer's line

Ne how the ground agast was of the light

(The Knight's Tale)

was to Lascelles Abercrombie a triumph of suggestion—the 'release, even from common words, of uncommon energy of meaning'. Bateson finds that the effect is simply that of a metaphor yoking fear and the insentient forest ground. The so-called suggested sense, as he sees it, is no mysterious exhalation that has a life of its own independent of the metaphor, but a clear cognizable meaning produced by the bridging of the 'semantic gap' between the two disparates brought together in the metaphor. This is a corollary of

74

SUGGESTION THROUGH METAPHOR

Bateson’s view that suggestion, understood as the opposite

of finite precise meaning, is a principle that has little relevance

outside Romantic poetry. Sanskrit poetics, on the other hand,

holds that suggestion is the highest mode of utterance in any

kind of poetry whatever and also, as we saw, that suggestion

is distinct from metaphor. It in fact recognizes a form of

suggestion that is based on metaphor (lakṣaṇādhvani)

(which is what we have been examining) and a form that

is based not on metaphoric but on primary meaning (ab-

hidhāmūladhvani). The former is also termed avivakṣitavācya

(‘where the stated meaning is not meant’), for metaphor

operates by replacing the primary with a secondary meaning.

The latter form of suggestion, which we dealt with in the

last two chapters, is called vivakṣitānyaparavācya (‘where the

stated meaning is meant and resolves itself into another—the

suggested—meaning’). This form consists largely, if not

entirely, in the suggestion of emotion, the stated meaning

being made up of the objective correlatives and the suggested

meaning being the emotion itself.

An altogether remarkable parallel insight makes Sir

Herbert Read formulate the very same distinction—between

metaphoric suggestion and suggestion of emotion.9 An

example he gives of the former is Tennyson’s line,

Now lies the Earth all Danaë to the stars.

(The Princess)

For the other kind of suggestion—‘simple statements devoid

of metaphor’ offering visual details which ‘have the power

of evoking the full reality in all its emotional significance’—

Read gives as an example Shelley’s lyric, ‘A widow bird

sate mourning for her love’, which presents grief (śoka)

suggestively through its objective correlatives.

SUGGESTION THROUGH METAPHOR 75

The Sanskrit theory of poetic metaphor, as we have seen, sets it in a three-tier meaning structure:

Vācyārtha. The literal meaning, thwarted by the incongruity between the word’s primary referent and the context. This is what happens in Hart Crane’s lines

The dice of drowned men’s bones he saw bequeath An embassy.

(‘At Melville’s Tomb’)

The incongruity stung Harriet Monroe into writing and asking Hart Crane how dice could bequeath an embassy.10

Lakṣyārtha. The metaphorical or secondary meaning, obtained by substituting for the primary referent a secondary referent related to it. We thus effect transfer of meaning, relate the vehicle to the tenor and paraphrase the poetic metaphor—fatally. Hart Crane did this to his own metaphors when he cleared Harriet Monroe’s puzzlement. ‘The dice of drowned men’s bones’ became ‘drowned men’s bones ground into little cubes by the action of the sea’; ‘bequeath an embassy’ became ‘washed ashore and offering evidence of messages about their experiences that the mariners might have had to deliver if they had survived’.

Vyangyārtha. Alongside of these two levels of meaning is a third—suggested meaning—through which the purpose or use of the metaphor becomes manifest. Hart Crane explained that he had employed the metaphor of dice as it was ‘a symbol of chance and circumstance’. This raises—or widens—the significance of the metaphor from the mere shape of the eroded bones to the whole ethos of Moby Dick’s world.

The Sanskrit account of metaphor is of twofold interest. First, it treats metaphor and suggestion not as the principles

of two types of image but as functions active in the same

image. Secondly, it answers partly the question we started

with: Why does metaphor seem to have more to it than the

transfer and why is it found offering a bonus of meaning

which turns out to be the poet's principal concern?

But the question is answered only partly. The Sanskrit

version of the metaphor phenomenon, by assuming a neat

removal of the literal meaning, has missed the fact that the

metaphor draws its sustenance from the obstinate vitality,

the unassimilable integrity, of the vehicle. In fact, suggested

meaning, which is very important to metaphor as seen by

Sanskrit analysts, most often springs from the incongruous

literal meaning. W. B. Stanford, who has offered, I think,

the best modern definition of metaphor, emphasizes the

fact that A and B (what we have been calling the primary

and secondary referents), though synthesized, 'retain their

conceptual independence'.11 A is not superseded; the incon-

gruity is not fully removed. Indeed if it were removed, if the

two terms of the metaphor existed in a state of fusion either

because the analogy was perfect or because (as in a dead

metaphor) A had been nicely absorbed into B, the metaphor

would suffer inanition. The vigour of the metaphor lies in

the very inadequacy of the similarity, in the unresolved or

partially resolved tension of disparates. 'Some similarity',

says I. A. Richards, 'will commonly be the ostensive ground

of the shift, but the peculiar modification of the tenor which

the vehicle brings about is even more the work of their

unlikenesses than of their likenesses.'12 Whether it is Aristotle's

'intuitive perception of the similarity in dissimilars', or

Coleridge's 'balance or reconcilement of opposite or dis-

cordant qualities', or Eliot's 'amalgamating disparate ex-

perience', the important factor is not the drawing together

of the two, but the continuing otherness of each. The vehicle and the tenor are brought close enough to each other to effect their mutual confrontation—and no closer. The vehicle not only maintains its distinctness and distance from, or even opposition to, the tenor but often becomes the more important—or even the only important—component of the metaphor. The vehicle is the creative part that the metaphor owes to what Freud called ‘the never failing source of all art’—the unconscious. It is the vehicle that links itself with other vehicles in the poem to form themes and patterns and to give the poem its coherence and depth. The vehicle, its original reference intact, its identity inviolate, reigns supreme within the metaphor—and its relation to the tenor and its congruity with the context are ever imperfect. The vehicle disrupts the statement and arrests attention. ‘To shock the audience by the violence and inadequacy of the analogy’ (as Martin Foss says, speaking of the ‘sick simile’)13 is the true function of metaphor. Its premises being what they are, the Sanskrit theory of metaphor, while admitting the fact of this shock, refrains from assigning any function to it and in fact provides for its quick resolution. Sanskrit poetics, like the classical poetry and the court drama from which it is evolved, is firmly committed to coherence and intelligibility. To us, however, the finding of a secondary meaning to remove the incongruity is but the establishing of rational meaning. The first tier (the disrupted statement) is the poem; the second (the meaning as it stands when the mess is cleared) is the paraphrase. Most of us would regard the latter as unimportant if not illegitimate. The shock is the thing.

The incongruity in what is being said pricks us into an awareness that more is meant than is being said, and we are

78 SUGGESTION THROUGH METAPHOR

laid open to the richer intimations that follow. The clash of

vehicle and context has the effect of damaging the walls

enclosing the meaning and it soon forces its way out,

creating conditions of flux and diffusion in which its less

precise and rigid elements, comprehensively called 'sug-

gestion', can operate freely. This, I think, is how the shock

works—it breaks the spell of logical discourse and renders

the reader's mind receptive to the circumambient suggestion.

Distinguishing between metaphor and transfer and linking

metaphor with equation and pregnancy, Empson in The

Structure of Complex Words14 refers to Gustav Stern's ideas:

Stern in Meaning and Change of Meaning distinguishes true

metaphor as intentional, emotive, dependent on context, and

accompanied by 'psychic resistance' . . . it seems to me that his

'intentional, emotive, and dependent on context', so far as they

are justified, all follow from his 'psychic resistance'. The thing

is felt as a sort of break in the flow, requiring interpretation,

exciting attention and perhaps other feelings.

Empson proceeds to redefine psychic resistance as resistance

not to a remote analogy but to a false identity. The fact

that by accepting the false identity you may 'fall into non-

sense' stimulates you into interpreting it, into 'typifying'

the vehicle (that is to say, concentrating on such character-

istics of it as are typical and essential for the case in hand)

so that 'pregnancy results. A transfer offers no such difficulty

and imposes no need for typifying. 'The rose of metaphor

is an ideal rose, which involves a variety of vague suggestions

and probably does not involve thorns, but the leaf of transfer

is merely leafish.' Empson sums up the whole process thus:

It seems to me that what we start from, in a metaphor as

distinct from a transfer, is a recognition that 'false identity' is

SUGGESTION THROUGH METAPHOR

79

being used, a feeling of ‘resistance’ to it, rather like going into

higher gear, because the machinery of interpretation must be

brought into play, and then a feeling of richness about the

possible interpretations of the word . . .

This version—that the resistance to the incongruity or the

danger of nonsense stimulates the reader to attend to the

richer interpretable meaning—corroborates our assumption

that the shock and the suggested meaning are related.15

Unless we keep the literal from the metaphorical meaning

and watch the disruptive action of metaphor on statement

(which is what yields the shock), we will not be able to

account for the emergence of the ‘intensive’ meaning. Allen

Tate’s essay on ‘Tension in Poetry’ ends with an interpreta-

tion of the ‘intensive’ meaning of Francesca’s words to Dante

who finds her and Paolo whirling in the winds of lust:16

Siede la terra dove nata fui

Sulla marina dove il Po discende

Per aver pace co’ seguaci sui.

(The town where I was born sits on the shore where the Po

descends to be at peace with his pursuers.)

The Po, says Allen Tate, is Francesca herself; the tributaries are the winds of lust which while pursuing her merge

in her, the sins merging in the sinner, even as the tributaries

merge in the river; the sibilants in the last line where the

tributaries are referred to complete the identification of

tributaries and hissing winds. This is brilliantly perceptive,

but by what process did all this meaning arise? ‘For although

Francesca has told Dante where she lives, in the most

directly descriptive language possible, she has told him more

than that,’ says Allen Tate, stopping short of the explanation

by millimetres. What has awakened us to the ‘more than

80

SUGGESTION THROUGH METAPHOR

that' is the disruption of the 'directly descriptive language'

by the metaphor for tributaries: seguaci. Seguaci is 'pursuers',

and Tate condemns Courtney Landon's translation of the

word as 'the streams that follow him'. This is paraphras-

ing, not translating—Landon offers the secondary meaning,

having clipped off the incongruity of the metaphor-word

( 'pursuers' ) before the English reader can see it and be shocked.

It is only when Tate has restored the word that the 'in-

tensive' meaning grows.

Whether it occurs in isolation as here, or in a cluster as

in Shakespeare or Shelley or Auden, or as a unit in a more

or less pure image-vocabulary as in Pound or in the early

Eliot or in Dylan Thomas, essentially, I think, a metaphor

breaks up the literal meaning and shocks you into attending

to the suggested meaning that accompanies or follows it.

It is impossible to test this hypothesis on a bunch or series of

metaphors and determine whether each metaphor delivers

its separate shock and introduces its own complement of

suggested meaning; or only contributes to the general dis-

solution of literalness and the creation of a climate of

evocation; or does neither. It is obviously easier to study

the behaviour of a metaphor when it is isolated in an en-

vironment of literal statement. An isolated metaphor will

then be found disrupting not only the literalness of the

statement which holds it but also the pervasive literalness of

the passage where the statement occurs. The first disruption

is the metaphor's essential function. The second, though it

obviously is no invariable component of a metaphor's

action, contributes to the shock as surely as the first does.17

(This phenomenon—the solitary metaphor—is the reverse of

what Day Lewis describes at some length in The Poetic

Image—a flow of images being arrested either by an explicit

general statement which gets them into perspective or by

a flat colloquial statement in the post-Symbolist manner

which shocks by its abrupt lowering of the emotional in-

tensity.)18

It is naturally in Wordsworth, in the later Eliot and in the

later Milton that one looks for passages of bare statement

where the lone metaphor can be seen at work. The first

sixty lines of the Third Book of The Prelude are sustained

statement, a Chaucerian description of the bustling Cam-

bridge scene. Tasselled cap, powdered hair, silk hose, college

kitchens and the Trinity clock are presented in plain or in peri-

phrastic language, and the metaphors such as there are are

of the unobtrusive faded variety. Suddenly the even surface

of statement is broken as a great metaphor shoulders itself

above it—the metaphor of Newton’s mind

Voyaging through strange seas of Thought, alone.

There cannot be a more adequate symbol for the scientific

imagination than the voyageur image, but there is more to

the metaphor than what it in itself is or means. The in-

congruity—the mind behind immobile stone being a wan-

derer—which breaks up the literal meaning of the state-

ment and the agitation it sets up in the prevailing plainness

of discourse make us sense a purpose behind the poet’s de-

parture from statement here, and we become aware of the

presence of peripheral meaning, the suggestion of a mystery

which the passage that follows describes, however fum-

blingly: Wordsworth’s mind, in solitude, turned first in-

ward and then outward, pushing its search to the very edge

of sanity, in an attempt to marry external forms to inward

states. The anticipatory suggestion of this phenomenon is

not generated but only triggered by the mariner metaphor. Even if the metaphor were not there, the words that precede it (Newton's 'silent face, / The marble index of a mind . . .' seen 'by light / Of moon or favouring stars') would carry the latent meaning that to Wordsworth the secret workings which the silence and repose of marble belie are the workings of his own mind, not Newton's. But this meaning would have remained latent and unapprehended, but for the metaphor.

Clearly in this part of the Third Book, Wordsworth is struggling to express certain operations of the mind which elude description and can perhaps only be shadowed faintly. The Newton passage is followed by another spell (some thirty-odd lines) of discursive literalness whose sole raison d'être seems to be to let itself be shattered by a curiously striking metaphor—Wordsworth's mind, when he had detached himself from Cambridge's dazzling show, 'into herself returning' and then with 'prompt rebound' regaining freshness and spreading itself 'with a wider creeping'—an athlete's or a ball's feat of instantly successive or near-simultaneous action and reaction, performed by the mind, perhaps to get uncoupled from 'the crowd, buildings and groves' and direct itself toward '. . . universal things . . . The common countenance of earth and sky'. The secret of this obscure operation—which must have been of some importance to Wordsworth, for he reverts to it twice in the next 110 lines—will let itself be spoken only through suggestion. Suggestion here is liberated by the recoil metaphor which, rising above a stretch of level matter-of-factness, introduces the athlete-(ball)-mind incongruity.

How Michael's spirit, reborn through his companionship with his ten-year-old son, survived the betrayal of that

companionship is the central meaning of Wordsworth's

Michael. But neither a neat propositional statement like this

nor the narration and description in the passage which

speaks of the rebirth can quite render the quality of that

experience. The overtones of meaning which uniquely

render it are set afloat by a double metaphor which intro-

duces the uneasy marriage of percept and affect and violates

the consistent literalness of the passage:

. . . from the Boy there came

Feelings and emanations—things which were

Light to the sun and music to the wind.

T. S. Eliot's play, The Elder Statesman, has a similar theme

—the redemption and regeneration of an old man through

the love of his daughter. There is the line spoken by Lord

Claverton:

I have been brushed by the wing of happiness . . .

The impact of the metaphor is violent, for apart from the

action of the bird image on the line itself (compare it with

the earlier unmetaphorical half-line:' . . . And now I feel

happy—'), the image bursts in at the end of a speech that,

for all its charge of feeling, is sternly austere in language, its

only two deviations from literalness being two metaphors

which are worn so thin as to be hardly noticeable as meta-

phors. In fact, it bursts in at the end of a play whose whole

tone and language, like those of its predecessors, are skilfully

stepped down to the low voltage of normal conversation.

The image defines the quality of the happiness Lord

Claverton has found—sudden, descended from above,

evanescent; but its more important office is to make us

aware of the fuller import of this moment of self-renewal

84

SUGGESTION THROUGH METAPHOR

and its far-ranging significance in terms of the whole play,

perhaps of the whole series of plays of which this is the last.

In Samson Agonistes the image (in its form a simile, not a

metaphor) of the 'two massie pillars' as mountains, which

tremble 'with the force of winds and waters pent', rears it-

self above the sustained plainness of the Messenger's curiously

objective reporting. The point of the image is obvious

enough, but occurring where it does, it lifts the meaning

from the referential to the evocative plane and attunes our

ears to the larger resonance of that climactic encounter of

the irresistible and the apparently irremovable—a widening

significance which takes us farther afield to Milton's own

predicament.

A fifth and last instance. The opening lines of Robert

Lowell's 'Memories of West Street and Lepke', which

record personal information in utterly unfigurative, almost

sterilized language, end by suddenly erupting into meta-

phor:

Like the sun she rises in her flame-flamingo infants' wear.

The immediate action of the triple metaphor is to specify

the colour of the garment, but it would be strange if that

were all when three vehicles—sun, flame and flamingo, all

equally at odds with a nine-month-old child—are massed in

a single line. They make a dry narration of facts end by

exploding into centrifugal meaning. The image for red

reaches out to the world, as yet unrevealed at this point, of

disintegration and despair which the succeeding parts of the

poem chart.

Not all metaphors, of course, work quite this way. All

that is claimed here is that often (if not always, as Sanskrit

poetics insists) a metaphor carries a load of suggestion and

SUGGESTION THROUGH METAPHOR

85

that in certain conditions its momentary disruption of

logical discourse quickens the reader’s sense of the suggested

meaning. This last happens only when the nature and the

situation of a metaphor permit it, but when it does happen,

it is a noteworthy effect—striking in itself and perhaps

significant also as an exploded model of a metaphor,

showing—something which we would otherwise not be

able to observe—how a metaphor essentially works.

6

Stating and Suggesting by Turns

What makes a poem switch from image to discourse or from discourse to image? One of the major achievements of recent critical theory has been to qualify the assumed polarity between the intensive manifold principle and the moral discourse principle and to recognize that symbol can be taught to lie down with discourse. Yet suggestion and statement giving rise to two opposed kinds of poem—Blake’s ‘Echoing Green’ and Goldsmith’s ‘Deserted Village’, to use Tillyard’s examples1—is an old concept; as old in fact as ninth-century Sanskrit poetics which distinguished between dhvanikāvya (suggestive poetry) and citrakāvya (representational poetry); and it is a concept that apparently still rules our thinking so that we mostly have either studies of symbolist poetry or studies, like Donald Davie’s,2 of ‘the poetry of urbane statement’. We do not seem to recognize that qualitative and rational progression as alternating modes within the same poem can be as significant a feature as the exclusive use of either. In fact, the alternation process can perhaps reveal better, because it does this contrastively, the cause and nature of either mode.

If poetry is not rhetoric, it cannot be all vision either; various a priori reasons have been assigned to show why it cannot. In the first place, poetry is words—and an art which

86

STATING AND SUGGESTING BY TURNS

87

employs verbal forms cannot altogether avoid discourse.

The Symbolist aesthetic never looked this basic linguistic

problem in the face. In the second place, although the

extreme theoretical position is that by image is meant not

this or that bit in a poem but the poem itself, yet most of us

will agree that revelation through symbol is, by its very

nature, intermittent and exists as high moments linked by

a process of social syntactic narration which acts, to use

Pound's phrase, as 'binding matter'. But these, I am afraid,

are merely general explanations of why a poem is necessarily

an articulation of symbolic and discursive components. It

would be more useful perhaps to take a few very familiar

poems—mostly anthology pieces—and note why there is in

each a transition from the one principle to the other.

Now it is fog, I walk

Contained within my coat;

No castle more cut off

By reason of its moat:

Only the sentry's cough,

The mercenaries' talk.

The street lamps, visible,

Drop no light on the ground,

But press beams painfully

In a yard of fog around.

I am condemned to be

An individual.

In the established border

There balances a mere

Pinpoint of consciousness.

I stay, or start from, here:

No fog makes more or less

The neighbouring disorder.

(Thom Gunn, 'Human Condition')

G

The first two stanzas are mythopoeic, but at the end of

the second, Gunn has already started explicating the import

of the myth. Having presented the vehicle of the metaphor

–the man in the coat walking through the fog–Gunn will

not let it do its work by itself; he must go on to state the

tenor. This probably means a gain in lucidity, but to know

what has meantime been lost, one has only to turn to

Gunn's poem 'Round and Round' which has the same

theme, existential isolation. 'Round and Round' shows the

interior of a lighthouse where the keeper has assembled his

minimum essentials–'A wife, a wireless, bread and brains'

–and these dance round with their faces turned toward the

centre. Nothing more is said. The manner of 'Human

Condition' recalls the explicitness of pre-Symbolist poetry

and is a conscious reaction against the tenor-muffling which

the Symbolist tradition used as a tool of suggestion. The

Symbolist symbol is the self-sufficient vehicle pointing to an

unnamed tenor; on the other hand in poetry such as that of

Wordsworth, Shelley, Tennyson or Arnold we occasionally

encounter moments when the myth having been presented,

the meaning is taken up for unfolding. What invariably

results is inanition of the image.

The effect is much the same or even worse when, as

happens in the following passages, the sequence is reversed

and the tenor is first offered, then the vehicle. Both passages

present the hermit image, which is no more than a coinci-

dence; both—and this is the point—preface it with a general-

ization.

Soul and body have no bounds:

To lovers as they lie upon

Her tolerant enchanted slope

In their ordinary swoon,

STATING AND SUGGESTING BY TURNS

89

Grave the vision Venus sends

Of supernatural sympathy,

Universal love and hope;

While an abstract insight wakes

Among the glaciers and the rocks

The hermit's sensual ecstasy.

(Auden, 'Lay your Sleeping Head, my Love')

Civilisation is hooped together, brought

Under a rule, under the semblance of peace

By manifold illusion; but man's life is thought,

And he, despite his terror, cannot cease

Ravening through century after century,

Ravening, raging, and uprooting that he may come

Into the desolation of reality:

Egypt and Greece, goodbye, and good-bye, Rome!

Hermits upon Mount Meru or Everest,

Caverned in night under the drifted snow,

Or where that snow and winter's dreadful blast

Beat down upon their naked bodies, know

That day brings round the night, that before dawn

His glory and his monuments are gone.

(Yeats, 'Meru')

The anticipatory affirmation that both these passages start with does greater damage to the image than a tenor-naming afterpiece could. Instead of thought and image getting formed together, the thought is pre-existent, and the image, however potent in itself, merely illustrates a proffered idea externally and is thus depressed to the level of a trope. This can be seen happening in another poem of Auden's:

About suffering they were never wrong,

The Old Masters: how well they understood

Its human position; how it takes place

90

STATING AND SUGGESTING BY TURNS

While someone else is eating or opening a window

or just walking dully along;

How, when the aged are reverently, passionately

waiting

For the miraculous birth, there always must be

Children who did not specially want it to happen,

skating

On a pond at the edge of the wood . . .

('Musée des Beaux Arts')

Eliot has a similar image of the world's unconcern and

normalcy on the day of Nativity, a strangely similar image

because Eliot too calls the birth 'hard and bitter agony',

though in a different sense:

Then we came to a tavern with vine-leaves over the lintel,

Six hands at an open door dicing for pieces of silver,

And feet kicking the empty wine-skins.

But there was no information, and so we continued . . .

('Journey of the Magi')

Eliot's image, with no prefatory exposition to impoverish it,

is free-standing and vital.

Where what is suggested is also stated, the suggested

meaning (as we saw in Chapter 4) loses its paramountcy and

what we then have is not suggestive poetry but 'the poetry

of subordinate suggestion' (gunībhūtavyaṅgya). Whether the

tenor is expounded before presenting the vehicle or after,

the effect is to debilitate the image; Sanskrit critics would

have no difficulty in classifying as gunībhūtavyaṅgya the

passages from Gunn, Yeats and Auden that we have ex-

amined. However, since each of these passages only expresses

the same thing twice over, stating and suggesting it by turns,

I am afraid they do not represent the process this chapter is

STATING AND SUGGESTING BY TURNS

91

centrally concerned with, which is the process of shifting

from the suggestive to the discursive mode or the other way

round, in response to a shift in content. An example of this

process is found in Yeats's 'Two Songs from a Play':

In pity for man's darkening thought

He walked that room and issued thence

In Galilean turbulence;

The Babylonian starlight brought

A fabulous, formless darkness in;

Odour of blood when Christ was slain

Made all Platonic tolerance vain

And vain all Doric discipline.

Everything that man esteems

Endures a moment or a day . . .

The whole of the first song and the first stanza, quoted

here, of the second song are propelled by a series of symbols

which, however pregnant and luminous, make no con-

cessions to the intellect. At the end of the intense sequence

we arrive at the direct assertion which the second stanza

opens with: 'Everything that man esteems Endures a mom-

ent or a day.' The metaphors that follow—the painter's

brush, the herald's cry, the soldier's tread—are so well-worn

as to be almost flush with the level of literal vocabulary;

they blend easily with the conceptualized nature of the

statement they support. The psychological rationale of this

transition to discourse is obvious. The discrete iridescent

symbols were a series of rapids and the stream must now

debouch into the commonality and comprehensibility of a

generalization and subside to a sedate pace. Something like

this is what Cecil Day Lewis talks about in the passage I

referred to in the last chapter, where he calls the discursive

92

STATING AND SUGGESTING BY TURNS

pause 'a point of vantage where we may rest a moment,

review the image-sequence over which we have passed, and

grasp its significance'.3

The device in Yeats's 'Two Songs' is, of course, different

from the built-in paraphrase I illustrated from Gunn. The

statement beginning 'Everything that man esteems . . .' is

not an explication of the import of the symbols; it is an

independent statement, arising from, but not re-expressing,

what the symbols intimate. The symbol-sequence was a

vision of the two great cycles of history; in the last stanza

the poem turns aside from this to affirm that each of man's

achievements perishes but not until it has used up something

of him that has gone into it.

Reverse the 'Two Songs' method, and you get a formula

whereby a spell of matter of fact is arrested by an abrupt

heightening. The even tenor of a flat description, narration

or piece of reasoning is broken when the voltage un-

expectedly rises, and a resonant moment arrives which

draws its power from the very stretch of referential exposi-

tion it helped to end. Masters of the literal, like Frost, are

equally masters of this effect. The first three stanzas, for

instance, of his 'Stopping by Woods on a Snowy Evening'

are a straightforward account of the woods, the evening

and the horse. The poem then instantly flowers into the

evocativeness of the last stanza which obviously owes its

force to the literalness of what went before.

The two passages from Yeats that we examined yield two

different explanations for a poem exchanging abstraction for

concretion or vice versa. Interestingly enough, there is a

third explanation available, again from Yeats, from the

Byzantium poems. John Bayley makes the point that the

use of self-contained symbols with 'a conversational armature

STATING AND SUGGESTING BY TURNS

93

in between’ (as in ‘Sailing to Byzantium’) is Yeats’s solution

for the Symbolist difficulty that an elaborate and unified

structure of symbols (as in ‘Byzantium’) tends to develop

allegorical reference.4 Further explanations for sandwiching

statement between symbols can be had, I think, from T. S.

Eliot’s poetry and criticism.

Eliot’s development after Ash Wednesday was away from

pure suggestion and towards the interleaving of suggestion

with statement. Four Quartets, which represents the con-

summation of this process, is an elaborate essay in oscilla-

tion between image and discourse.

There is no end of it, the voiceless wailing,

No end to the withering of withered flowers,

To the movement of pain that is painless and motionless,

To the drift of the sea and the drifting wreckage,

The bone’s prayer to Death its God. Only the hardly

barely prayable

Prayer of the one Annunciation.

It seems, as one becomes older,

That the past has another pattern, and ceases to be a mere

sequence—

Or even development: the latter a partial fallacy

Encouraged by superficial notions of evolution,

Which becomes, in the popular mind, a means of

disowning the past.

(“The Dry Salvages”)

This is typical of the way the non-discursive and discursive

modes succeed each other in Four Quartets, and for an ex-

planation of why any poem should be founded on such

alternation we can turn to Eliot himself—to his criticism.

Didn’t he say, ‘. . . the poet, at the back of his mind, if not

as his ostensible purpose, is always trying to defend the

94

STATING AND SUGGESTING BY TURNS

kind of poetry he is writing, or to formulate the kind that

he wants to write . . . ?5 The date of ‘The Dry Salvages’ is

1941. In the Ker Memorial lecture he gave in 1942, Eliot is

found trying to defend the kind of poetry he was writing or

to formulate the kind that he wanted to write. He argues

that ‘in a poem of any length, there must be transitions

between passages of greater and less intensity’, and that ‘this

test of the greatness of a poet is the way he writes his less

intense, but structurally vital, matter’.6 As Poe saw, texture

implies brevity; conversely, length implies structure, and by

1942 Eliot had travelled all the way from pure texture to

texture worked into structure. The Quartets are elaborately

structured, with the ‘less intense matter’ and suggestive writing

deployed in a well-defined pattern that repeats itself in each

poem. Eliot’s ‘less intense matter’ is, of course, the same con-

cept as Pound’s ‘binding matter’ and John Bayley’s ‘intellec-

tual armature’.

A second reason Eliot produces elsewhere for the to-and-

fro movement between the prosaic and the poetic is that

length implies variety and variety implies smoothness of

such movement. ‘It is indeed necessary for any long poem,

if it is to escape monotony, to be able to say homely things

without bathos, as well as to take the highest flights without

sounding exaggerated.’7 This is true enough, but Eliot has a

third and strikingly perspicacious explanation for the image-

discourse alternation. The transitions, he says, ‘give a rhythm

of fluctuating emotion essential to the musical structure of

the whole’.8 This extremely significant concept is developed

fully in his lecture, ‘Poetry and Drama’, where he demon-

strates how the change from mere verse to poetry and back

in the opening scene of Hamlet yields a musical pattern that

alternately ‘checks and accelerates the pulse of our emotion’.9

STATING AND SUGGESTING BY TURNS

95

This can happen, as he points out, twice in the space of four lines:

So have I heard, and do in part believe it.

But, look, the morn, in russet mantle clad,

Walks o'er the dew of yon high eastern hill.

Break we our watch up.

Of course, the disposition of statement and suggestion is a vastly more intricate business in poetic drama where, as Eliot points out, the musical design should parallel the dramatic structure and a given utterance should express either character (where it does not meaningfully transcend it) or situation. In non-dramatic poetry, where these complicating factors are absent, the musical design tends simply to reproduce the archetypal pattern of repose and excitement.

Apart from the analogy of music, I think the fact that emotion, by its very nature, continually ceases and reappears dictates the dialectic of discourse and image. The transition from ebb to flow of emotion is accompanied by a transition from statement to suggestion (or the other way round, as the case may be), because emotion does not tolerate discourse. Frost’s ‘Birches’ is a useful model demonstrating this process. There is first a bare description of the way the boy climbs the birches and alights from the swinging branches; apart from the point made that the game flatters the boy’s sense of mastery, the account is free of all tinge of emotion.

The line

So was I once myself a swinger of birches

signals a sharp change in direction and interest. The poet swings round from the boy to himself, from fact to feeling,

96

STATING AND SUGGESTING BY TURNS

from the denotative to the connotative manipulation of language—and the boy’s birch-swinging which was an objectively recorded game continues as his own birch-swinging, an image now, suggestive of the mutual tension within him of the escapist and homing emotions.

7

Statement Poetry

Is there such a thing as statement poetry at all? When Tillyard says, ‘All poetry is more or less oblique: there is no direct poetry’,1 or when Wimsatt says, ‘Poetry is never altogether, or even mainly, “poetry of statement”’,2 if all they mean is that there is an element of suggestion in the most pedestrian discursive poem (even as there is an alloy of discourse in the most intensely symbolist poetry), that is something one would have no difficulty in granting, since it cannot be claimed for either kind of poetry, or indeed for any kind, that it occurs in a pure state. We talk about ‘statement poetry’ only because the difference between Cowper’s The Task and Mallarmé’s L’Après-midi d’un Faune is so great that we cannot pretend that they are the same kind of writing.

But it was on altogether different grounds from these that the existence of statement poetry was denied by Susanne Langer. If by statement poetry, she says, is meant poetry which offers direct propositions, then it is not poetry at all, as propositions belong to science and are alien to art. Now this is largely a matter of how you choose to define ‘poetry’; a definition which excluded Chaucer and Dryden and several parts of Wordsworth would not have much to recommend it. Mrs Langer does not indeed exclude them—her manner of attacking statement in poetry is to engulf it (as the amoeba engulfs its prey) and assimilate it into her theory. According

97

to this theory of art as symbolic form, when a poet states a thought directly, his stating of it is an illusion and what has really happened is that the thinking of that thought—an occurrence in the thinker's personal history, an experience lived and felt by him—has got presented. But this is something Lascelles Abercrombie had said twenty-seven years before Suzanne Langer. Speaking of the ‘Time hath, my Lord, a wallet’ speech in Troilus and Cressida, he says the real matter of the passage is not the thought it presents, ‘because it is not presented to us simply as thought, but as the finely emotional and subtly allusive experience of an individual mind thinking—of, precisely, Shakespeare's Ulysses thinking’.3 There is, no doubt, something in this, but a theory which assumes that every stated thought in poetry is a semblance beneath which is the reality—the process of thinking and feeling—frustrates itself by its very inclusiveness. By this token, for instance, instead of identifying the following lines as pedestrian verse offering a mere proposition about human treachery, we would have to salute them as symbolizing the poet Rochester's entertainment of the thought. The point about the lines, of course, is that underneath the deadening abstraction, the experience, if there is any, has so feeble a pulse:

Which is the basest Creature, Man or Beast? Birds feed on Birds, Beasts on each other prey, But Savage Man alone, does Man betray: Prest by necessity, they Kill for Food, Man undoes Man, to do himself no good. (‘A Satyr against Mankind’)

But we ought not really to be talking about individual lines and passages when our concern in this chapter is with

whole poems. We are dealing here with statement poetry as

a genre, and neither with statement as a mode alternating

with the suggestive mode within a poem (see the last

chapter) nor with statement—stated meaning, that is—as a

component of meaning, serving as a base for the other

component, i.e. suggested meaning (see Chapter 4). State-

ment poetry—poems that are merely expository, descriptive

or narrative—has a long history and is still a living form—it

is an unbroken tradition stretching from Gower or farther

back all the way down to Betjeman. Sanskrit criticism calls

this genre 'representational or pictorial poetry' (citrakāvya),

distinguishing it from 'suggestive poetry' (dhvanikāvya)* and

from 'the poetry of subordinate suggestion' (gunībhūta-

vyaṅgya). The distinction between poetry where the sug-

gested meaning is secondary and poetry where it is para-

mount is not easy to sustain—at any rate, not easy to demon-

strate. But the notion that poetry aiming merely at repre-

senting or stating does exist can be a useful corrective to the

heresy that statement in poetry is an illusion or a kind of

prestidigitation.

Citrakāvya is a pejorative term—Sanskrit criticism regards

statement poetry as an inferior, if not wholly illegitimate,

kind. This, I think, is an equally useful concept. Donald

Davie, whose expositions of statement poetry are the best

available yet, concedes that to find 'purity of diction',

which he equates with 'strength of statement', 'we should

look not among our great poets but among our good ones'.4

Susanne Langer, on the contrary, argues that the distinction

between statement poetry and oblique poetry 'rests on a

difference of technical means rather than of poetic excel-

• 'All Art that is not mere story-telling, or mere portraiture, is symbolic.'

(Yeats, Ideas of Good and Evil)

lence',5 and that since Blake's means and Goldsmith's

means, although different, are means to the same end, the

idea of a special standard to judge Goldsmith by is artificial.

Of course, once we get outside the terms of Mrs Langer's

theory of statement as semblance, we see easily enough that

a value judgement is implicit in the very term 'statement

poetry'. And this has never been clearer than today. Aside

from interludes like the Movement, the commitment to

indirection and semantic richness as essentials of the best

poetry is almost universal in our age. However, our age is

also known to have a weakness for description and analysis,

at the expense of evaluation, in criticism, and there is always

the danger that our perception of degrees of excellence will

be deadened if the notion that the status of statement poetry

is essentially different is not insisted upon.

On grounds very different from Mrs Langer's it can be

made out that discursive and suggestive are not absolute

categories and that whether a poem is suggestive or the

opposite depends on how you read it. You could, if you

tried hard enough, read Cowper as though he were Swin-

burne and Swinburne as though he were Cowper. This

would no doubt be an exercise calling for an effort and

feasible only within limits, but it is feasible, and I could try

to establish its feasibility by means of a demonstration. But

this has already been done for us. In Interpretations, the con-

cluding four lines of 'Among School Children'—for range

and richness of meaning, probably the most remarkable

single passage in English poetry of this century—are treated

by John Wain as merely referential and disposed of in the

following brief paraphrase:

... the confused relationship of matter and spirit. Even blossoming

and dancing are accompanied by this confusion; a chestnut-tree

has a massive trunk, cool green leaves, and delicate blossoms; at

which point is one most in touch with its essential identity?

Again, a dancer is the embodiment of the dance; without the

tangible, moving body, the dance would not exist; nevertheless

it is a perceptible thing in itself. This question is brought up, but

it is not the function of the poem to propound a solution . . . 6

In the very next chapter, G. S. Fraser demonstrates how

an immense wealth of meaning can be read into Denham’s

well-known four lines on the Thames which, their metaphors

notwithstanding, are explicit and direct. Now Fraser, not

Wain, typifies the reading habit of our time. Trained by con-

temporary writing to be perpetually tuned-in to unspoken

meanings, the twentieth-century reader tends to respond to

‘The Deserted Village’ or even ‘An Essay on Criticism’ in

this way. This does work—but not all the time and not

beyond a point, because a successful poem makes you read

it the way it wants to be read and effectively rejects any

other way as alien to its nature. In other words, you

cannot, except as a tour de force, read Cowper as though

he were Swinburne nor Swinburne as though he were

Cowper.

The case for conceding that statement poetry can exist in

its own right is, I think, strong. It has, in fact, so existed

since Chaucer or as far back as one cares to push—and as an

unbroken tradition, central and prestigious during certain

periods and tending to withdraw to the periphery during

others. Indeed a whole poem structured on narration,

description or exposition and employing lucid, image-free,

self-contained statement as its language has been such a

familiar form in the past that it is unnecessary to offer

examples, although some poets have been more closely

identified with this mode than others, and one thinks here

102

STATEMENT POETRY

of Chaucer, Gower, Dryden, Pope, Johnson, Goldsmith,

Cowper, Crabbe and Clare. The list stops short of

our own century—and so it would even if it were more

detailed and comprehensive. Today a long discursive

poem, when one does get written, is a distinguished

exception.

While talking thus about the fortunes of statement poetry,

one has got to be very wary of the occupational hazards

of literary history so impressively illustrated in the fate of

Eliot’s theory of how discourse and image (or thought and

feeling, if you like) got dissociated in 1650 or thereabouts.

It is not arbitrary, I think, to take it that Edgar Allan Poe’s

attack (in The Poetic Principle) on the idea of the long poem

was the starting point, or at least was the first sign, of the

decline of statement poetry. The attack was motivated by

the premises of the aesthetic he was enunciating; his theory

of suggestiveness could be erected only on the dead body

of the long poem, because (although a few long poems in

the Symbolist mode have indeed been achieved) intensity

presupposes brevity. Poe’s objections to statement and to

length were soon refined and developed into the central

assumptions of the Symbolist poetic. But quite apart from

Poe and the Symbolists, the popularity of Palgrave’s Golden

Treasury (1861/69) was creating a taste for the intense short-

flighted lyric, and the novel was fast replacing the long

narrative poem as fireside family reading. In the manifestos

of the first wave of the ‘revolution’, Mallarmé, building on

Schopenhauer’s ideas, had defined the change as the replacing

of the language of naming and description with the language

of suggestion; in the manifestos of the second wave, Hulme,

building on Bergson’s ideas, defined the change as the re-

placing of language ‘as an abstract process’ with the language

of images serving as ‘a compromise for a language of intuition’.7 This movement away from directness and explicitness and indeed from logical and social values has ruled out the use of statement as a poem’s normal voice and demoted it to a secondary rôle either as coarse connective tissue or as an alternative mode to switch to occasionally. Further, in our own times, texture in poetry and close reading in criticism have concurrently grown in importance. It is difficult to decide which of the two is the cause and which the effect, but both the trends have clearly worked against the extended use of statement in poetry.

Nevertheless, statement poetry still persists although as a somewhat isolated tradition, and an examination of a few samples of this surviving mode will probably turn out to be a better aid to an understanding of its decline than my timid excursion into literary history has been. Also, the differentia and rationale of statement poetry are more easily discernible in its present-day models than in earlier ones, because it is not the normal kind today and its features acquire greater definition against the reigning mode.

A work that can serve as a point of departure for this enquiry is Robert Lowell’s Life Studies—not the whole volume, but the third part, forming a separate sequence of poems, after which the book itself is named. Life Studies has been described as a breakthrough in Lowell’s work in several ways, but what is most strikingly novel in it is the laconic featureless language of case-histories in which it records events in Lowell’s own life and in the lives of other Lowells and Winslows. If his well-known early poem ‘The Quaker Graveyard in Nantucket’ is stylistically his Paradise Lost (the comparison is Randall Jarrell’s), then Life Studies can well be his Paradise Regained, representing a movement

104

STATEMENT POETRY

away from verbal richness and indirection and toward austere bareness of statement. ‘The Quaker Graveyard’ speaks of Warren Winslow’s death thus:

The wind’s wings beat upon the stones,

Cousin, and scream for you and the claws rush

At the sea’s throat and wring it in the slush

Of this old Quaker graveyard where the bones

Cry out in the long night for the hurt beast

Bobbing by Ahab’s whaleboats in the East.

In the late 'forties, when Lowell mourned the death of his grandfather, Arthur Winslow, it was still in the same language:

and the ghost

Of risen Jesus walks the waves to run

Arthur upon a trumpeting black swan

Beyond Charles River to the Acheron

Where the wide waters and their voyager are one.

(‘In Memory of A. W.’)

The congestion of imagery here, the coiling eloquence, the myths from antiquity, religion and literature and the incantatory phraseology and rhythm raise the language of these passages to an extraordinary degree of complication and intensity of meaning. Life Studies, on the other hand, exhibits the terminal result of the process of dilution and sterilization to which Lowell subjected his language in order to reduce it to transparent referential statement. Lowell’s passages on death in Life Studies—first his father’s death and then his mother's—are a strange contrast to the early elegiac moments:

STATEMENT POETRY

105

Father’s death was abrupt and unprotesting.

His vision was still twenty-twenty.

After a morning of anxious, repetitive smiling,

his last words to Mother were:

‘I feel awful.’

(‘Terminal Days at Beverly Farms’)

Mother travelled first-class in the hold,

her Risorgimento black and gold casket

was like Napoleon’s at the Invalides . . .

In the grandiloquent lettering on Mother’s coffin,

Lowell had been misspelled LOVEL.

The corpse

was wrapped like panetone in Italian tinfoil.

(‘Sailing home from Rapallo’)

The language of Life Studies is dead-pan matter-of-fact re-

porting, relieved by a very occasional heightening; and the

bleakness of the language is matched by the muted verse.

‘To a Cisatlantic ear’, says F. W. Bateson, ‘they [the Life

Studies poems] are rhythmically formless, prose disguised as

verse by chopping it up into short lines.’8

Given a year,

I walked on the roof of the West Street Jail, a short

enclosure like my school soccer court,

and saw the Hudson River once a day

through sooty clothesline entanglements

and bleaching khaki tenements.

Strolling, I yammered metaphysics with Abramowitz,

a jaundice-yellow (‘it’s really tan’)

and fly-weight pacifist,

so vegetarian,

he wore rope shoes and preferred fallen fruit.

106

STATEMENT POETRY

He tried to convert Bioff and Brown,

the Hollywood pimps, to his diet.

Hairy, muscular, suburban,

wearing chocolate double-breasted suits,

they blew their tops and beat him black and blue.

('Memories of West Street and Lepke')

This reads like the transcript of a recording of casual

coffee-bar reminiscing; but the shame and despair within,

the corruption and disruption of the psyche, wait in the

background, and the lines have a menacing aspect. This

does not, however, alter the fact that the style represents an

experiment in drastic paring down to starkly simple state-

ment. The unique idiom of Life Studies stands out the more

sharply because Lowell's more recent work—Near the Ocean

(1967), for instance—represents a return to symbol-making

and to tighter and more obvious rhythms. The concluding

stanza of 'Fourth of July in Maine' is typical:

We watch the logs fall. Fire once gone,

we're done for: we escape the sun,

rising and setting, a red coal,

until it cinders like the soul.

Great ash and sun of freedom, give

us this day the warmth to live,

and face the household fire. We turn

our backs, and feel the whiskey burn.

The raison d'être of Lowell's excursion into statement

poetry is, I think, clear enough. His evolution up to the

point represented by Life Studies was from bardic intensity

to dramatic speech and thence to confessional sincerity; and

obeying the logic of this development, his language changed

from image and incantation to the variety and ease of the

speaking voice and then to plain statement. Now the ‘I’ of

the dramatic monologue denotes only its protagonist and if

it ever has any relation to the poet, it is to the poet with the

‘mask’ on. This ‘I’ is, therefore, a franchise for the imaginary

in content and the imaginative in style. On the other hand,

the ‘I’ of the avowedly confessional poem stands, quite

literally, for the person who wrote it and commits him to

accurate presentation of autobiographical fact and, equally,

to a language appropriate to unrefracted transmission of in-

formation. To state this distinction in even simpler terms,

the poetic symbol and the confessional first person singular

are opposed principles—the former involves suggestiveness,

the latter involves statement. Of course, artistic sincerity

is more than literal veracity; artistic presentation of auto-

biographical material obviously does involve selection and

re-ordering with a view not merely to form but indeed to

the total truth. And the confessional ‘I’ can in certain cases

admit mythopoeia as, for instance, in the confession of

mystic experience; it can, I suppose, even admit self-

dramatizing to some extent. All the same, it would be, by

and large, right to say that symbol and ambiguity are the

language of essentially dramatic poetry like Eliot’s and are

alien to the purposes of a confessional poet like the Lowell

of Life Studies. The Life Studies poems were originally ac-

companied, in the same volume, by a prose autobiographical

sketch, ‘91 Revere Street’, whose content they largely echo.

If Lowell wanted to project in material form the concomi-

tance of confession and prosaic statement, he couldn’t have

used a better method.

The assumption I have offered of the concomitance of the

two has, of course, got to be tested on other samples of

confessional poetry. The most obvious sample from modern

108

STATEMENT POETRY

poetry would be D. H. Lawrence’s work, the poems which

he classifies as ‘personal’ as distinguished from ‘fictional’.9

In his ‘personal’ poetry, of which the most impressive

single body is the sequence, Look! We have Come through!,

Lawrence discards the mask—or (to use his own metaphor)

doesn’t put his hand over the mouth of the Demon in him

and speak for him but lets him say his say.10 Lawrence not

only describes this sequence as personal but insists unequivo-

cally that it should be treated as a factual autobiographical

record of the years 1912 to 1917 in his life:

Even the best poetry, when it is at all personal, needs the

penumbra of its own time and place and circumstance to make it

full and whole. If we knew a little more of Shakespeare’s self and

circumstance how much more complete the Sonnets would be to

us, how their strange, torn edges would be softened and merged

into a whole body! So one would like to ask the reader of Look!

We have Come through! to fill in the background of the poems, as

far as possible, with the place, the time, the circumstance. What

was uttered in the cruel spring of 1917 should not be dislocated

and heard as if sounding out of the void.11

Here the objection in present-day criticism to the mask-

less poet inside the poem being identified with the man who

wrote it is anticipated and effectively disarmed. Besides this,

in Look! We have Come through! Lawrence is already using

free verse, claiming for it that it ‘is, or should be, direct

utterance from the instant, whole man’, utterance that

‘rushes out without artificial foam or artificial smoothness’.12

In a different context and on a different ground, Lawrence

describes ‘stark, bare, rocky directness of statement’ as the

essence of poetry today.13

Yet, for all these indications to the contrary, confession in

direct language is precisely what Look! We have Come

through! is not. By ‘direct statement’ Lawrence doesn’t seem

to have meant a transparent unfigurative style, a language

reduced to the consistency of prose. Rather, the ‘directness’

he preaches denotes immediacy (as opposed to finish) in

utterance, the pulsating plasm as opposed to the perfected

unageing gem. This ‘momentaneity’ is the presiding prin-

ciple of Look! We have Come through! Lawrence, of course,

did practise stark, bare, rocky statement, but as the vehicle

not of confession but of lively exact description, so that it

is frequently encountered in the collection, Birds, Beasts, and

Flowers:

She is large and matronly

And rather dirty,

A little sardonic-looking, as if domesticity had driven

her to it.

Though what she does, except lay four eggs at random

in the garden once a year

and put up with her husband,

I don’t know.

She likes to eat.

She hurries up, striding reared on long uncanny legs

When food is going.

Oh yes, she can make haste when she likes.

('Lui et Elle')

This language of naturalistic portraiture is a far cry from the

charged language of metaphor and resonance to which

Lawrence was roused in the confessional passages of Look!

We have Come through!:

Then the sweeping sunshine of noon

When the mountains like chariot cars

Were ranked to blue battle—and you and I

Counted our scars.

110

STATEMENT POETRY

And then in a strange, grey hour

We lay mouth to mouth, with your face

Under mine like a star on the lake,

And I covered the earth, and all space.

('History')

Look! We have Come through! is an extended epithalamium shot with direct self-revelation, but it certainly does not employ the unechoing explicatory language of statement poetry.

John Betjeman's Summoned by Bells, which appeared a year after Life Studies, is also a full-length exercise in confessional statement poetry, but both its confession and its statement are noticeably different. Look! We have Come through! is, in Lawrence's words, 'a biography of an emotional and inner life',14 and so is Life Studies. Confession, if that is the word, in Summoned by Bells is not a voice from the buried cellars of being—it moves, in the wake of Betjeman's bicycle, at the open-to-the-skies, sun-drenched level of 'those winding lanes of meadowsweet and umbelliferae'. Betjeman, who tells how he in his youth began in gladness, cannot ever claim that thereof came in the end despondency and madness. The sensibility he reveals is sophisticated but never demonic or tortured. His mode of statement too is different. Unlike Lowell's clipped taut sentences, Betjeman's are full and flowing; they are overt, explicit, completed, lucid—almost overpoweringly lucid.

It cannot be claimed that the discursiveness of Summoned by Bells was dictated by its confessional nature—discursiveness is Betjeman's native mode anyway. As Philip Larkin said, for Betjeman 'there has been no symbolism, no objective correlative, no T. S. Eliot or Ezra Pound, no rediscovery of myth or language as gesture, no Seven Types or Some Versions'.15 A very brief analysis here of how all

Betjeman's instincts combine to make easy urbane discourse his inevitable medium might help in isolating the major determinants of statement poetry in general. To begin with, for Betjeman, communication is the first imperative; and the kind of rapport with the reader which his art demands as the very condition of its being establishes itself not at the subliminal level at which suggestion largely works but at the level of rational consecutive exposition. That places him firmly in the traditional line of present-day poets—and this in fact is the second determinant of statement poetry operative in Betjeman's case. What drew him to the 'Tooled leather, marbled paper, gilded edge' of old books and the lonely, heavily restored St Ervan's church in a Cornwall hollow must have also drawn him to traditional modes of poetic utterance with their emphasis on unfoldedness and socialness. Thirdly, Betjeman's devotion to the lineaments of the external landscape—manors and churches, 'lanes in beechy Bucks', Wiltshire cottages, the 'neo-Tudor shops' of suburbia, Cornwall beaches, and human beings ranging from Sarah the maid ('with orange wig and horsey teeth') to Maurice Bowra himself—gives his writing a strong bias toward down-to-earth description of an almost Chaucerian liveliness and gusto; and statement is its tool. And in Betjeman's case, the common disease of statement poetry, bathos, is averted by the prophylactic action of the mock-heroic vein and the readiness to laugh at himself.

Apart from illuminating thus some of the primary determinants of statement poetry, Summoned by Bells raises a question which, I think, is central to any examination of statement and suggestion. The prefatory note to the poem describes Betjeman as having gone 'as near prose as he dare'. Indeed, the poem is an exercise in brinkmanship of con-

112

STATEMENT POETRY

siderable virtuosity, but Betjeman several times treads the

edge too confidently and is pitched into prose:

Those were the days when Huxley’s Antic Hay

Shocked our conventions, when from month to month

I rushed to buy The London Mercury,

And moved from Austin Dobson on to Pope.

This is iambic prose and hasn’t even got the light-verse

effect which redeems the Homeric catalogue of second-hand

books elsewhere in the poem:

Mason’s Works

(But volume II is missing), Young’s Night Thoughts,

Falconer’s Shipwreck and The Grave by Blair,

A row of Scott, for certain incomplete,

And always somewhere Barber’s Isle of Wight;

The antiquarian works that no one reads—

Church Bells of Nottingham, Baptismal Fonts

(‘Scarce, 2s. 6d., a few plates slightly foxed’).

The question we would be tempted to ask on encounter-

ing this kind of ‘poetry’ has already been asked—by Betjeman

himself in the prefatory note: ‘Why is this account of some

moments in the sheltered life of a middle-class youth not

written in prose?’

And why, we may ask, is T. S. Eliot’s play The Confidential Clerk not written in prose?

eggerson. You will wish to obtain confirmation

Of this interesting discovery, Mr. Kaghan,

By putting your adoptive parents in touch

With Mrs. Guzzard. It’s for them to confirm

That they took you, as a child, from Mrs. Guzzard,

To whom, it seems, you had first been entrusted.

This language, although clearly not quite identical with that

of Betjeman's memoirs, prompts the same question as his, and if Betjeman, having put the question, did not answer it explicitly, Eliot has explained at some length, chiefly from the dramatist's angle, how statement poetry differs from prose. But Eliot's conception of the rôle of statement in poetic drama is developed from his conception of the rôle of statement in poetry in general, and we must look at the latter first. It is set forth in the lecture on 'The Music of Poetry' where Eliot makes the point that in a long poem suggestion is necessarily confined to certain moments when the suggestiveness or what he calls the 'allusiveness which is in the nature of words' is activated and that the rest of the poem is less intense matter; the structure of the poem rests on the disposition of the richer among the poorer words. This is a familiar enough notion; what is significant is Eliot's further assumption that the poorer words, which account for the bulk of the poem, are so vital to its structure that the way the poet manages them is a test of his greatness. In fact 'it may be said that no poet can write a poem of amplitude unless he is a master of the prosaic'.16 The chief aspect of this mastery is the ability 'to say homely things without bathos'—a complementary skill to that of taking 'the highest flights without sounding exaggerated'.17 Eliot adds a diachronic gloss on this binary view of the language of poetry—literary history, as he sees it, patterns as an alternation of periods when the language of poetry develops in the direction of elaboration and complication, with periods when it returns to the supple rhythms and direct vocabulary of spoken everyday language and catches up with its changes; exaggeration of the former tendency makes suggestiveness artificial and of the latter makes statement pedestrian.

114

STATEMENT POETRY

It was these general assumptions about statement poetry that Eliot applied to the problems of drama when in the late 'thirties he started experimenting in a new kind of play. Reacting both against high-faluting poetic drama and naturalistic prose drama, and discarding the language of his own first major play, he proceeded to fashion the kind of statement poetry for drama that is typified in the Confidential Clerk passage we noted above. The language of conventional poetic drama was, of course, repugnant to contemporary life which was the stuff of his plays from The Family Reunion onwards. Prose, on the other hand, was equally alien to his purposes—when moments of emotional intensity made poetry the only valid language, the transition from prose to poetry could jolt the present-day audience (which, unlike the Elizabethan, regards poetry as many degrees removed from language in life) out of that unawareness of the medium which is the first condition of dramatic response. Statement poetry seemed to be the answer. It was, for one thing, poetry 'with nothing poetic about it . . . poetry so transparent that we should not see the poetry, but that which we are meant to see through the poetry';18 secondly, it would enable him 'to bring poetry into the world in which the audience lives and to which it returns when it leaves the theatre';19 thirdly, statement poetry would be a medium 'in which everything can be said that has to be said'.20 It is true that when Eliot spoke of a new medium based on a 'proper modern colloquial idiom' as he called it, he meant not only a language composed of the unfigurative, unevocative vocabulary of educated conversation but also a metric made up of the natural stresses of contemporary speech which could replace blank verse now grown remote from spoken English and somewhat stiff in the joints. We

are in this chapter strictly concerned only with the first of

these two components—what Raymond Williams, speaking

of the dialogue in The Cocktail Party, calls 'statement, of a

deliberate lucidity, and with the minimum of imagery and

evocation'.21 Yet it seems to me that the answer to Betjeman's

question, Why is this not written in prose? lies in the results

achieved by Eliot's dramatic verse which, while as a rule

carefully stopping short of the poetical, accomplishes a

variety of jobs that prose cannot attempt. The verse is so

self-effacingly unobtrusive that we are not aware that it has

gone to work on us—particularly in the theatre, where,

unless we have read the play, we have no means of know-

ing whether we are listening to verse or prose. The impact

on us of the high points of intensity in the plays does bear

out Eliot's claim that the ascent to these is rendered smooth

for the modern audience when the ground level is verse, not

prose. Apart from this, any scene in the plays would show

that features of rhythm, which even in the most speech-

based and loose verse are more precise and clearly marked

than they would be in prose, can be made to serve as a

gesture of individuality, to mark a rise in tension, or to

colour a variation in tone. In addition, the basic pattern of

three stresses, which is audible all the time as a subdued

undertone and, though not consciously attended to, is

nevertheless something the ear learns to expect, can, when

departed from, signal a shift in the situation or a turn in the

argument. If nothing else, the unceasing muffled ticking of

the three-stress line can insinuate, amid the medley of

voices, a suggestion of unity. Verse, as Eliot claimed, is

never 'merely a formalization, or an added decoration',22

and the potentialities of low-voltage verse, which Eliot's

plays constantly worked towards realizing, can validate the

116

STATEMENT POETRY

flattest statement poetry and give it an identity and status distinct from those of prose.

'Your trouble is not sticking to the subject,'

Pike said with temper. And Dick longed to say,

'Your trouble is bucolic lack of logic,'

But all he did say was, 'What is the subject?'

'It's whether these professions really work.

Now take the Doctor—'

In its directness, clarity and closeness to conversation, the dialogue in Frost's 'From Plane to Plane' (the lines quoted above are typical of the poem) shares the same language with the dialogue in The Confidential Clerk. The language of The Confidential Clerk we can easily enough recognize as something Eliot voluntarily chose for a specific genre, but in the case of Frost, there is the absurd phantom, not fully exorcised yet, of the New England farmer who can do no more than versify and naïvely moralize. For the dialogue in 'The Death of the Hired Man', Frost draws upon the same plain language of everyday speech:

'No, but he hurt my heart the way he lay

And rolled his old head on that sharp-edged chair-back.

He wouldn't let me put him on the lounge.

You must go in and see what you can do.

I made the bed up for him there to-night.

You'll be surprised at him—how much he's broken.

His working days are done; I'm sure of it.'

'I'd not be in a hurry to say that.'

'I haven't been. Go, look, see for yourself.

But, Warren, please remember how it is:

He's come to help you ditch the meadow.

He has a plan. You mustn't laugh at him . . .'

STATEMENT POETRY

117

This is explicit narration of how Mary interceded with her husband for the human waif, Silas. But what this passage overtly states, an earlier passage in the poem has presented by image:

Part of a moon was falling down the west,

Dragging the whole sky with it to the hills.

Its light poured softly in her lap. She saw

And spread her apron to it. She put out her hand

Among the harp-like morning-glory strings,

Taut with the dew from garden bed to eaves,

As if she played unheard some tenderness

That wrought on him beside her in the night.

It is obvious from this that when Frost is not stating, he can wield the opposite method with equal assurance; but the point is that this is the poem’s only passage in the language of image. Indeed, imagery is not, by any means, Frost’s customary language. There are in fact two Frosts—a fact which both his critics and apologists seem to forget. Cleanth Brooks once said that ‘Frost’s themes are frequently stated overtly’23 and that his best poetry, however, ‘exhibits the structure of symbolist-metaphysical poetry’24—the distinction I am trying to draw here is different. There is, on the one hand, the Frost of ‘Directive’, ‘Design’, ‘Acquainted with the Night’, ‘Sand Dunes’, ‘Come In’ and like poems, who, whether his language be pregnant image or simple-seeming statement, is concerned to find a voice for his specifically twentieth-century sensibility, complex, in-growing and loneliness-haunted. There is then the other Frost who is content, through easily accessible verbal structures, to portray in realistic detail a scene, an animal, a human figure; or to develop a situation by coherent narration; or simply to present and unearned generalization. ‘The Run-

away', for instance, is a frankly naturalistic vignette of a young colt agitated by its first snow. 'Two Tramps in Mud-Time' paraphrasably affirms the wisdom of uniting love of work and need of work. 'From Plane to Plane' employs the sun not as a many-meaninged symbol but pointedly as a logical illustration of two concepts not poetically fully validated: the first, 'extrication' or detachment from professional activity; and the second, the donor's shyness of gratitude. 'The Death of the Hired Man' is an episode simply told and a somewhat Galsworthian portrait of failure; and, except for the very competent blank verse, it does not employ any device that would not be equally legitimate in a prose 'character' or short story. These poems are successful on their own plane, the plane of statement; to burden them with unstated significance and to despise, as some of Frost's apologists do, the reader who is content with their literal meaning is to thwart the proper response to them by forcing a reaction that is appropriate to a different kind of poem where Frost lets his inwardness speak either through symbol or through pregnant prosaic statement.

If we seek the origins of the discursive vein that distinguishes a great deal of Frost's work, we will find that although his statement poetry, at its plainest, subsists, in comparison with Betjeman's, at a different—or, if you like, higher—level, yet the same determinants of statement poetry are active in both. The compulsion to communicate successfully, to be, in the best sense of the word, popular, was strong in Frost. Secondly, his interest in places and persons, in seasons and creatures, in their real selves and for their individual sakes, was equally strong, so that a horse, a cow in Frost, unlike the members of Ted Hughes's menagerie, are a horse, a cow, neither more nor less. Thirdly, although

the idiom he fashioned for himself is a very individual instrument and its chattiness can at times be a cloak for less obvious intimations, yet even while working toward effects of this kind and certainly at other times, Frost is content to operate within the framework of traditional modes of utterance and be discursive and intelligible. In fact, his commitment to economy of metaphor and allusion raises his language to a higher degree of lucidity of affirmation than one associates with traditional writing.

However, to equate unmetaphorical unallusive language with statement, as we have sometimes done, is to oversimplify, at least in certain cases. Pope’s metaphors, for instance, offer a univocal logical enunciation:

For wit and judgment often are at strife,

Tho’ meant each other’s aid, like man and wife.

’Tis more to guide, than spur the Muse’s steed;

Restrain his fury, than provoke his speed.

(‘An Essay on Criticism’)

On the other hand, a poem like George MacBeth’s well-known ‘Report to the Director’ can affect the impersonal colourlessness and workaday precision of civil service writing or speech and yet, by its omissions and implications, develop layers of sinister meaning:

The infusion

Was one of the smoothest I’ve seen. Evacuation

Very decent. An infinity of freshness

In a little diffusion of bitter carbolic. Rather sweet.

It took about fifteen minutes to get the stories,

And not much mess; they had to scrub the channel

To clear some vomit, otherwise all O.K.

This kind of ‘statement poetry’ or understatement poetry,

beneath its carefully preserved discursive exterior, attempts suggestion without using the more familiar tools of suggestion like image, evocation or ellipsis. Much of the work of the Movement poets too, despite their professed contempt for ‘the myth-kitty’ and for incantation, and despite their definition of a poem as a structure of events or arguments, is sophisticated non-statement of this sort, fraught with a quantity of suggested meaning.

8

Suggestion as a Classical

Method

'It is an obvious truth of romantic poetry', says John Bayley

in The Romantic Survival, 'that exact words usually "sug-

gest" far more powerfully than vague ones.'1 To be fair

to Mr Bayley, he makes this observation while examining

certain words in Dylan Thomas's poetry, which do manage

to be both exact and suggestive; and Dylan Thomas is

clearly a neo-romantic. Yet if I were generalizing, I would

be inclined to say the very opposite: that it is an obvious

truth of Romantic poetry that vague words usually suggest

far more powerfully than exact ones. It is the strength of

much Romantic poetry that, to borrow Wimsatt's words,

the 'shadowy suggestion of abstractive categorizing' forbids

us to descend to 'the substantive level'.2 Shelley's 'Champak

odours', the nineteenth-century Thomson's 'wine of love'

and Swinburne's 'Grief with a glass that ran'3 are typical

Romantic words; light in referential content, they float high

above the specifically qualificative plane. They are merely

centres of semantic radiation. Of course, all Romantics are

not alike. As T. S. Eliot says, 'The bird of Wordsworth

"breaking the silence of the seas" remains; the swallow of

"Itylus" disappears'.4 Wordsworth's cuckoo does refer to the

object, while Swinburne's swallow is little more than the

word's sound and associations.

121

122

SUGGESTION AS A CLASSICAL METHOD

Yet the difference is only one of degree; and if Swinburne’s

swallow differs qualitatively from any bird, it is from Eliot’s

own plover. The bent golden-rod and the whirling plover

of Ash Wednesday are, in the first place, solidly denotative and

form an exact notation of the New England coastal scene.

In the second place, their suggestiveness is not unlimited

but points firmly to a specific inward experience. The differ-

ence is seen no less in the use of less visually concrete images.

The time symbol, for instance, works in radically different

ways in the following two passages:

Before the beginning of years

There came to the making of man

Time with a gift of tears;

Grief with a glass that ran . . .

(Swinburne)

Time and the bell have buried the day,

The black cloud carries the sun away.

Will the sunflower turn to us . . .

(Eliot)

Eliot’s, I think, is an experiment which, though half a cen-

tury old now, is still of great interest and significance. It

also links itself, across the intervening centuries and despite

great differences, with another experiment.

Suggestion is a device characteristic of Romantic poetry;

some, like F. W. Bateson,5 would even say it is peculiar

to Romantic poetry and has no general application. The

meaning that is suggested in Romantic poetry is diffuse,

imprecise; the quality that enables a word or image to sug-

gest such meaning is, to borrow Wimsatt’s words again,

‘the dreamy abstractness, the suffused vagueness of revery’.6

Hulme’s banner of revolt against Romanticism was the

SUGGESTION AS A CLASSICAL METHOD

123

classicist principle of the accurate, the precise, the definite.

Eliot seems to have taken over from the Romantic tradition

its technique of suggestion and adapted it to classical values,

evolving thus a new mode of suggestiveness that blends

precision and control with range of reference and indirection.

There is a time for the evening under starlight,

A time for the evening under lamplight

(The evening with the photograph album).

('East Coker')

We soon find that the poet has decided for us that in reading

these lines we do not tend 'as when a pebble is dropped in a

pool, to watch meanings opening out in rings',7 but rather

to 'determine the precise degree of evocation of particular

figures'.8

The danger in substituting regulated richness of meaning

for freely eddying meaning is that you can slip into one-

to-one allegorical correspondence. This seldom happens in

Eliot. There is, for instance, a carefully preserved difference

between the three leopards of Ash Wednesday and the three

beasts of the Inferno. The leopards have ordered associations

of terror and beauty which are worked into the poem's

structure, but they do not admit of equation with a concept.

Eliot's vehicles have a directed significance, but nowhere are

their tenors named.

In a sense, of course, 'controlled suggestion' is not a dis-

covery to be credited to Eliot. It is rather the consumma-

tion of a tendency that inheres in the Symbolist method.

One has only to look at pre-Symbolist romantic poetry or

at automatic writing of any period to realize that before

symbolism became a faith and a technique, the dreaming

poet could think alogically and mythopoeically, and his

124

SUGGESTION AS A CLASSICAL METHOD

symbols, being spontaneously created and unconsciously

used, were non-rational and non-referential and possessed of

portmanteau meaning. Once, however, the poet becomes

aware that he is using symbols, he cannot use a symbol any

more without it symbolizing (at least for him) something.

The predicament of the Symbolist poet is just this: that a

consciously used symbol is a contradiction in terms, in that

it necessarily develops the finiteness of reference that is fatal

to a symbol. No poet who knows that he is being sugges-

tive can help suggesting something more or less definite!

To look this dilemma in the face, as Eliot did, is not to

slide from Symbolism into allegorizing or tenor-naming but

to recognize that willed suggestion necessarily resolves itself

into controlled and channelled release of multiple meaning.

It is significant that Lascelles Abercrombie, although a

critic in the Romantic tradition, thought of suggestion as very

much a contrived thing. He was led to this position not via

the Symbolist impasse but by his notion of suggestion as

secondary meaning. In the once influential essay, Principles

of Literary Criticism, he says: ‘In fact, literary language differs

from ordinary language precisely by the conscious and de-

liberate use in it of powers additional to the force of gram-

matical meaning . . . Literary art, therefore, will always be in

some degree suggestion . . . A very large part of literary skill

consists in vividly liberating for its effect on imagination just

that particular secondary meaning in words which is not only

appropriate to the immediate occasion, but which will make

the occasion come to life in the reader’s mind.’

In order to watch ‘controlled suggestion’ at work and

compare its mechanics with those of Romantic suggestion,

we can place a stanza from the hymn in Prometheus Unbound

alongside a stanza from the ‘Little Gidding’ lyric.

SUGGESTION AS A CLASSICAL METHOD

125

Life of Life! thy lips enkindle

With their love the breath between them;

And thy smiles before they dwindle

Make the cold air fire; then screen them

In those looks, where whoso gazes

Faints, entangled in their mazes.

Who then devised the torment? Love.

Love is the unfamiliar Name

Behind the hands that wove

The intolerable shirt of flame

Which human power cannot remove.

We only live, only suspire

Consumed by either fire or fire.

There are four images or terms—fire, love, life and breath

—common to the two stanzas; the way they work in each

is significant. In Shelley each term acts as a source of semantic

exhalation; and the terms are not encouraged to impinge

actively on one another. In Eliot, the same terms are found

moving by continual back-and-forth reference to one

another; they are placed in a specific if ambivalent relation-

ship to each other of tension and confirmation, and they

simultaneously extend and define each other’s meaning.

‘Fire’ in Shelley’s line is meant to have unlimited resonance.

But as Yvor Winters once pointed out, maybe rather too

hard-headedly, our reaction to the word ‘fire’ would gener-

ally depend on whether the word as used on the occasion

relates to a fire on a hearth, in a furnace, or in a forest!9

This defining power of context is active in Eliot’s cor-

responding line. The two groups of associations that ‘fire’

has—fire as desire that destroys and fire as purgatorial fire

126

SUGGESTION AS A CLASSICAL METHOD

that refines and redeems—are disposed there contrapuntally,

and the second group is echoed and enriched by earlier lines

in the lyric and indeed elsewhere in the poem. Certain

associations of the word are thus selectively established, and

the other associations are either excluded or muffled. The

significance of ‘love’ is similarly defined by the references to

divine grace in the context and earlier. ‘Live’ and ‘súspire’,

operating in a way radically different from that of Shelley’s

‘life’ and ‘breath’,10 are in tension with ‘consumed’ in the

next line, and this relationship determines their meaning.

Indeed, the whole of Four Quartets is a system of interlocked

symbols that release and regulate one another’s meaning,

demonstrating that suggestiveness can be reconciled with

classical precision . . .

Hung instantly upon the eyelashes, smote

The lips, got shattered next on the high breasts,

Ran down the stomach’s stages—and at last

The first drops of rain sank in her navel.

(From Kumārasambhava, an epic poem by

Kālidāsa, the best known classical Sanskrit

poet)

This chart of the raindrops’ course is suggestive of

Umā’s beauty; each feature of it, long eyelashes, full lips,

firm breasts—if these had not been such, the raindrops

would not have behaved thus on each—answers perfectly

to the traditional ideal. There is also a further suggestive-

ness here working through tension: the hinted loveliness of

flesh is part of the picture of Umā in the erect attitude of the

stern ascetic discipline whose object was the hermit god

Śiva’s love. Nothing can, on the face of it, be more different

and distant from this than Eliot’s picture in Ash Wednesday:

SUGGESTION AS A CLASSICAL METHOD

127

The broadbacked figure drest in blue and green

Enchanted the maytime with an antique flute.

Blown hair is sweet, brown hair over the mouth blown,

Lilac and brown hair . . .

Kālidāsa’s lines are suggestive through their indirection,

Eliot’s through their evocativeness. Kālidāsa’s picture is

complete as far as it goes, though it has a significance

added to it; Eliot’s is just three glimpsed details which call

up the whole vision. Yet one has only to compare the two

passages with Shelley’s

And wherever her aëry footstep trod,

Her trailing hair from the grassy sod

Erased its light vestige, with shadowy sweep . . .

(‘The Sensitive Plant’)

to realize that the difference between Kālidāsa and Eliot is

less fundamental than the difference between Shelley and

Eliot. The suggestive power of Shelley’s lines is the power

of n, and his images work by width and vagueness of refer-

ence. Kālidāsa’s lines and Eliot’s lines, on the other hand,

have a suggestiveness pointing to a finite unstated meaning

which, in the case of the pair of passages compared here,

happens to be the same: the beauty of the flesh seen against

a spiritual ascent that leads beyond it. The sense of such

beauty is expressed suggestively through its objective cor-

relatives—the full lips and brown hair over them blown.

Using ‘a set of objects’ to ‘evoke’ a ‘particular emotion’

(the phrases are from Eliot’s well-known pronouncement

on the objective correlative) is a principle valid for all art.

What makes Eliot’s formulation distinctive is its insistence

that the objects should be sensuously concrete and the

128

SUGGESTION AS A CLASSICAL METHOD

emotion-meaning particular. In much Romantic poetry

neither the suggestive object nor the suggested meaning has

definition. What Eliot was involved in was thus an ex-

periment in taming the wild energies of suggestion and

teaching it to be servant to the classical values of precise

saying and finite meaning. This is what one might call

'classical suggestion', and there had been an experiment in it

in a very much earlier and very different body of classical

poetry and criticism, but Eliot clearly was not aware of this.

He knew of ancient Indian philosophy—the Upaniṣads, the

Bhagavad-Gītā and the Buddha's sermons—but as late as in

1955, writing to Nimai Chatterji, an Indian correspondent,

he spoke of 'the author of Vibhava [sic], which I have not

read'.11 That he had no contact with Indian aesthetics only

makes Eliot's experiment the more interesting.

It would, of course, be as absurd to liken Eliot's poetry to

Kālidāsa's as to liken it to Virgil's. Kālidāsa's work—ex-

pressive of a simple untroubled vision, given to discursive

narration, opulent in imagery—would indeed seem to be all

that Eliot's is not. The point I am making is only that Eliot's

poetry (of which his theory of the objective correlative was

either a defence or an advance formulation)12 and Kālidāsa's

poetry (which, with other poetry of its kind, formed the

basis of the Sanskrit theory of the objective correlative),

while differing in all else, agree in adapting suggestion to

classical values.

Forthwith the aśoka's boughs burst into shoot and blossom—

He wouldn't wait till women's ankleted feet had touched him.

(Kumārasambhava)

These lines from Kālidāsa, on an untimely spring, can be set

beside Eliot's lines on 'midwinter spring':

SUGGESTION AS A CLASSICAL METHOD

129

Now the hedgerow

Is blanched for an hour with transitory blossom

Of snow, a bloom more sudden

Than that of summer, neither budding nor fading,

Not in the scheme of generation.

('Little Gidding')

The blossom of snow is part of a symbol, unusual rather

than conventional, objectifying the 'stirring of the dumb

spirit'. The aśoka blossom is similarly a detail in what is

an extended objective correlative (a description of the spring,

filling a sizeable portion of the canto) for 'the heart's heat'

(to borrow Eliot's words) that disturbed the ascetic god

Śiva when Umā appeared before him. Yet, while Eliot's

image for the disturbance is the flaming ice acting on the

soul's sap, Kālidāsa's image is the new-risen moon acting

on the sea. The two images indicate the two widely dis-

parate methods of objectification used by Eliot and Kālidāsa.

Kālidāsa's objective correlatives are explicit, uncomplicated,

sedately familiar when they are not merely ingenious. The

aśoka flower, like all flowers, is a traditional uddīpanavibhāva

('stimulant-correlative') for love, and women's tread making

the aśoka tree flower is a conventional idea. Nothing can be

farther removed from the difficult condition or concept Eliot's

lines are struggling to indicate.

Kālidāsa had the advantage of being born into a cultural

closed system with a large fund of shared responses to ob-

jects so that within it the sensuous correlates of emotions

for its members had a large measure of uniformity; he was

also blest with membership of a poetic tradition which

probably had already worked out a larger number of

emotion-object relationships than most traditions do and

had erected them into conventions. It was otherwise with

130

SUGGESTION AS A CLASSICAL METHOD

Eliot. The Elizabethans, and in a different way the Augustans,

had the notion of an ideal order in nature and society to

serve them as a frame of reference. The Romantics did not

inherit a framework, but if they had inherited one, their

commitment to the absolute sufficiency of the subjective

principle would have ruled out any acceptance of external

authority. Eliot's reaction against Romanticism was to

evolve the concept of 'an ideal order' formed by 'the existing

monuments' of 'the whole of the literature of Europe from

Homer'. In submitting to the authority of this Christianized

Graeco-Roman culture and the literary tradition it sup-

ported Eliot saw the answer to the anarchy and centrifugality

of his own age and to all that repelled him in Romantic

art. Their devotion to the ego and to the spontaneous over-

flow and the absence of an objective 'myth-kitty' made the

Romantics employ arbitrary and personal symbols whose

suggested meaning worked not by being widely and pre-

cisely recognized but by blurred reference and infinite

diffusion. Eliot substitutes for this a method of allusion

and quotation yielding symbols which, being drawn from

tradition, combine range and reference with concreteness

and precision and, having always been part of a system, are

more amenable to control and direction. This surely is how

the lady in the white gown of Ash Wednesday differs from

the damsel with the dulcimer in Kubla Khan and the Arab

maiden in Alastor. The images Eliot finds from English and

European literature, Indian sacred literature, fertility myths

and Christianity are easily identifiable within their own

cultures which are accessible to all those who care, and be-

cause these images are public—because they indeed were the

currency of thought and feeling in the past—they do not

share the irresponsibility and unpredictability of private

symbols and are non-volatile and tractable where private symbols tend to fade far away and dissolve. When Eliot does use personal images, he ensures that they are so disposed as to illuminate each other—or else he links them immediately with known literary or religious lore and universalizes their reference. Even if this is not done, a private image (according to Eliot), if it is ‘consciously concrete’, will always become ‘unconsciously general’. Conversely, in the hands of the Romantics the hallowed images from mythology and nature they sometimes used tended to get filled with a personal meaning that transformed them beyond recognition.

The canto from which I quoted the lines about the aśoka tree has a massing of the objective correlatives of love—the mango and karnikāra blossoms, the cuckoo sweet-throated from eating the mango sprouts, the thirsty bee, the creepers bent with bunches of flowers heavy as breasts. Umā’s girdle of flowers slipping from her waist, the glow of perspiration on the kimpuruṣa women’s bodies. Many of these are conventional and stylized, maybe even stereotyped, but exactly because they are the objects that tradition has attached to the sexual emotion, they operate as precise and potent suggestors of it. Indeed, an objective correlative is, by definition, traditional—it is ‘a specified, figuratively fortified, and permanent object’¹³ that literary tradition has evolved for the corresponding emotion. Also, an objective correlative works best in a classical climate—in an established cultural situation where responses to the ‘objects’ are more or less uniform and stabilized. The objective correlative is thus an essentially classical device. Even C. K. Stead—who is committed to the view that Eliot’s theory and earlier practice recognize ‘unconscious process’, and not ‘conscious direction’, as the essential element in poetic composition—

132

SUGGESTION AS A CLASSICAL METHOD

identifies the objective correlative with the poem’s willed structure which is the classical principle active in the poem.14 (Eliot’s belief, on Mr Stead’s showing, is that what is most truly poetic in a poem has an inspirational or automatic origin; but Mr Stead accepts Eliot’s view that ‘there is a great deal, in the writing of poetry, which must be conscious and deliberate’15—like, for instance, the task of editing what has risen from below-conscious levels, or the technical tasks of craftsmanship. This is what I meant when I referred to Eliot’s experiment as one in ‘taming the wild energies of suggestion and teaching it to be servant to the classical values of precise saying and finite meaning’.) There has been to show that ‘Eliot’s literary position cannot be called classical’.16 Mr Le Brun discovers a Romantic basis for the objective correlative—the objective correlative, as he sees it, is the work of art itself, a direct and immediate presentation of the poet’s perfectly fused thought and feeling. The poem, being thus a sensuous equivalent to an exceptionally integrated consciousness, embodies a special kind of insight. But this is closer to Bergson, Cassirer and Susanne Langer than to Eliot who meant by the objective correlative not the whole poem seen as a symbol but an individual image or set of images in it, such as, to reproduce Eliot’s own example, the sensory details projecting the sleep-walking Lady Macbeth’s state of mind.

Clearly then, the objective correlative as conceived by Eliot is classical in nature and function. And suggestion, though generally a Romantic method, is, among other things, the business of evoking an emotion through its objective correlative. In this aspect, then, suggestion is a classical method. By developing the possibilities of suggestion in this direction and

SUGGESTION AS A CLASSICAL METHOD

133

by attempting control and command of what used to be regarded as mindless and riderless, Eliot not only put suggestion to an unusual use but put himself in unusual company -the company of some dead masters whom he had not known.

9

Suggestion or Statement?

The Case of Wordsworth

Every time what looks like a symbol in Wordsworth beck-

ons me, I think of Patricia Beer’s sonnet about him:

Winter was not a symbol, nor was Spring,

Nor was the corpse that floated to the air

After a week of water, nor the wing

Of the December star pinned to the mere

By a child’s skate. His dawns were literal,

His ghosts did not melt from the ice of darkness

But froze on into the sunshine. Guilt was real

And the stern mountain had no other likeness.

A lake was something that could drown him, though

It danced, he said. The river had no voice

Although it sang. He knew too well the plan

By which the world shared neither grief nor joy

And stood for nothing else, but really was

The wet and dry, the hot and cold of man.1

Much has been written about Wordsworth’s commitment

to the actual—a given thing was of interest to him because it

was what it was in itself. This is true enough, but a poet can

affect factualness on the surface while he transmits extra-literal

meaning at other levels. And even if his literalism is proved

to be genuine and pre-emptive of any intended symbolism

134

whatever, it need not invalidate the extended meanings that

offer themselves when contact is made between poem and

reader. This is theoretically the case. In practice, however,

an object that the poet has elected to present naturalistically

may not accommodate the kind of response that would

treat it as a symbol, and then the quest for wider mean-

ing soon enough runs into rough weather. David Perkins's

attempt, for instance, to study the cottage girl in 'We are

Seven' and the six-year-old in the 'Immortality Ode' as

examples of the Romantic child symbol is blocked by the

fact that both the children are presented as children in such

realistic detail that to perceive anything at the 'level of symbol

is made difficult by what is perceived at the more insistent

level of fact, and Perkins is obliged to say, ' . . . there is a

question whether the poem can be said to use a symbol'.2

That sea, and mist, crag and torrent were symbols to

Wordsworth in his life does not necessarily mean that they

function as symbols in his poetry. Obviously, these are two

different meanings of the word 'symbol'. An object or

event that has been to a poet a felt symbol in life can be

quite properly introduced by him in a poem to exist as a

symbol in the latter sense. But a poet who, in doing so,

describes it as a symbol and declares what it symbolizes

is thereby interpolating the element of explication that pre-

cludes it from acting as a poetic symbol. The winds thwart-

ing winds and the black drizzling crags on the slope of the

Alps, which are described as 'symbols of Eternity', and the

moon above Snowdon which is explicated at some length as

'the emblem of a mind', quite obviously do not even begin

to be poetic symbols.

There is another sense in which the winds on the Alpine

slopes cannot be symbols in the poem-they, like the loud

dry wind that roared in his ear as Wordsworth hung above the raven's nest, are a real enough agent or factor of the mystic experience, not a fictional counter of it as Dylan Thomas's 'rocketing wind [that] will blow The bones out of the hills' clearly is. When an object or event is conceived as a component of reality and set in the cause-consequence chain, its possibilities as a symbol are aborted. Once framed in a rational context, it 'makes sense' in terms of itself and of the context, and the compulsive need for a further meaning does not exist. A symbol, on the other hand, that is not cause-linked becomes independent of the plot and free to carry many-layered meaning. There is thus a qualitative difference between 'The torrents shooting from the clear blue sky' on Wordsworth's Alps, on the one hand, and on the other, the piano set up by Madame Somebody on Rimbaud's Alps and the 'flamingoes on the mountaintops at dawn' in Karen Gershon's 'Swiss Morning'.

David Perkins—I return to him, as his is the most recent considerable quest for symbols in Wordsworth—examining the well-known 'Magnificent the morning' passage, speaks of the obscure symbolic suggestion in the lines, arising, he thinks, from the contrast between the sunrise on the mountains (images of spirit) and the dawn in the meadows and fields (images of the human).3 One feels, however, that the passage is no more than the record of an important actual experience and owes its beauty to effects other than symbolism. The high tones of the peaks and the subdued light and the stir in the valley are what any Hollywood camera, innocent of symbolism, would have picked out in a contrastive sweep in the normal course. The occasion is Wordsworth's dedication of himself to 'solitary study' and 'meditative peace'; the dawn, encountered in just the con-

THE CASE OF WORDSWORTH

137

ditions that would make its impact powerful, must have in

actual fact caused (and maybe, being a dawn, also symbolized

to Wordsworth) the commencement of a new life of single-

minded submission to nature. In the poem it is far from a

symbol—it is no more than a literal dawn.

The assumption that suggestion rather than statement is

Wordsworth’s central mode would have to be tested on

those passages in his work which deal with experience that

is most obstinately resistant to discursive presentation—

on the great nodal passages, that is, of The Prelude which

refer to his instants of illumination. Surely here, if nowhere

else, Wordsworth would be driven to the use of symbol?

The apocalyptic passages that constitute the poetic core of

The Prelude follow a more or less uniform pattern. There is

first an explicit description of the situation that triggered the

experience. The description ends abruptly and (one suspects)

before it is complete, and a veil descends. And the next

thing is a sonorous passage of apostrophe and philosophic

musing. Between the two is the central hiatus—the moment

itself, presented neither obliquely nor direct, indeed pre-

sented not at all, so that when the passage ends the secret is

still inviolate. The mystical experience is led up to, hinted at,

talked about, but nowhere is it presented either by suggestion

or by statement. Consider, for example:

. . . the sky seem’d not a sky

Of earth, and with what motion mov’d the clouds!

The mind of Man is fram’d even like the breath

And harmony of music. There is a dark

Invisible workmanship . . .

(II. 349–53)

138

SUGGESTION OR STATEMENT?

But huge and mighty Forms that do not live

Like living men mov'd slowly through my mind

By day and were the trouble of my dreams.

Wisdom and Spirit of the universe!

Thou Soul that art the Eternity of Thought!

(I. 425-9)

... and I stood and watch'd

Till all was tranquil as a dreamless sleep.

Ye Presences of Nature, in the sky

Or on the earth! Ye Visions of the hills!

(II. 488-91)

And all the answers which the Man return'd ...

Ended in this; that we had cross'd the Alps.

Imagination! lifting up itself

Before the eye and progress of my Song ...

(VI. 521-6)

What we have in the text is the context and the doctrinal comment—and both, though they cannot express the arcane experience, keep talking about it. The experience is affirmed, explained, commented upon in the poem, but it does not happen within its language. The moment is not embodied in the poem by 'sensuous re-creation'. This has resulted in a general feeling that while the reality of Wordsworth's mystical communion is undeniable, his way of expressing it is that of a spectator. Archibald MacLeish sees this as the difference between Rimbaud's visionary poetry and Wordsworth's: 'Where Wordsworth asserts the moment of insight and tells us what the insight was—“Our birth is but a sleep and a forgetting”—Rimbaud evokes the actual images in which the insight exists.'4

Talking about the thing, expounding it propositionally,

THE CASE OF WORDSWORTH

139

or naming it, is Wordsworth’s way not only with the clearly

ineffable states but with all states of emotion.

Expressing an emotion is not the same thing as describing it.

To say ‘I am angry’ is to describe one’s emotion, not to express it.

The words in which it is expressed need not contain any reference

to anger as such at all. Indeed, so far as they simply and solely

express it, they cannot contain any such reference …5

The words are Collingwood’s, but they might well be

Mammaṭa, the eleventh-century Indian critic, describing

svaśabdavācyatā (‘stating by its own name’) which is the first

in his list of rasa-doṣas (‘defects in the presentation of poetic

emotion’). I wonder what Mammaṭa would have said about

the statements in Coleridge’s ‘Dejection’ where the emotion

is identified and labelled and its symptoms recorded with

clinical accuracy and directness. Certainly, Wordsworth’s

instinct seized upon statement as peculiarly suited both to his

literalness and to his easy informative egotism.

Any study of Wordsworth’s discursive method quickly

develops into a study of his language—and here one has to

compare, a process which, Wordsworth’s status being what

it is, involves comparing the greater with the lesser. Here

are two pairs:

(1) A motion and a spirit, that impels

All thinking things, all objects of all thought,

And rolls through all things.

(“Tintern Abbey”)

The force that through the green fuse drives the flower

Drives my green age; that blasts the roots of trees

Is my destroyer.

And I am dumb to tell the crooked rose

My youth is bent by the same wintry fever.

(Dylan Thomas, ‘The force that through the green fuse’)

140

SUGGESTION OR STATEMENT?

(2)

... Magnificent

The morning was, a memorable pomp,

More glorious than I ever had beheld.

The Sea was laughing at a distance; all

The solid Mountains were as bright as clouds,

Grain-tinctured, drench'd in empyrean light.

(Prelude, IV. 330–5)

The point of one white star is quivering still

Deep in the orange light of widening morn

Beyond the purple mountains: thro' a chasm

Of wind-divided mist the darker lake

Reflects it: now it wanes: it gleams again

As the waves fade, and as the burning threads

Of woven cloud unravel in pale air:

'Tis lost! and through yon peaks of cloudlike snow

The roseate sun-light quivers.

(Shelley, Prometheus Unbound, II. i. 17–25)

One realizes how generalized the words 'motion' and

'spirit' are and how little work, apart from the alliteration,

'magnificent' 'memorable' and 'glorious' do. To describe

words of this kind (as R. A. Foakes does in The Romantic

Assertion) as 'value words' betokening man's highest hopes

and achievements or to argue (as G. C. Clarke does in his

Romantic Paradox) that Wordsworth uses words like 'form'

and 'shape' to express his sense of the essentially equivocal

nature of perception does not alter the fact that their sug-

gestion potential is feeble. Wordsworth did often plump for

the non-thingy general word which, however 'lucid', can

only state where the sensuous image potently suggests.

To take just one such word. 'Motion' occurs six times

within the space of 133 lines in Book I of The Prelude:

THE CASE OF WORDSWORTH

141

sounds

Of indistinguishable motion . . .

(330–1)

. . . and with what motion mov'd the clouds!

(350)

With measur'd motion, like a living thing . . .

(411)

a breath

And everlasting motion . . .

(430–1)

spinning still

The rapid line of motion . . .

(481–2)

All the passages have reference to Wordsworth's flashes of insight. Returning to the theme seventy lines later, he returns to the same word:

Those hallow'd and pure motions of the sense

(551)

The word had already been used in the parallel statements of 'Tintern Abbey':

And even the motion of our human blood

Almost suspended . . .

A motion and a spirit, that impels

All thinking things . . .

Now it is clear that, whatever the word means in these lines, it does not mean 'movement'. When Wordsworth means movement plain and simple, he says movement:

The coarser pleasures of my boyish days,

And their glad animal movements all gone by.

142

SUGGESTION OR STATEMENT?

It is equally clear that ‘motion’ does not mean the same

thing in these passages. In expressing different if related

experiences, not all of them equally capable of expression

except obliquely through a suggestive image, Wordsworth

must have found that ‘motion’, by its blurred outline and

imprecise reference, was a useful alternative to image.

When the stars threw down their spears,

And water’d heaven with their tears . . .

(‘The Tyger’)

This is Blake’s visionary sky. Confronted with a like

apocalyptic transfiguration overhead, Wordsworth finds

generality and tautology sufficient:

the sky seem’d not a sky

Of earth, and with what motion mov’d the clouds!

We can, of course, treat motion (as opposed to becalming

or arrest) and spirit (identical originally with wind, breath,

life) as archetypal images, as Maud Bodkin in fact does.6

Or we might treat ‘motion’ as a word which Wordsworth

uses in an idiosyncratic way, making it a carrier of a special

or private meaning. What is more likely, however, is that

the word, so inclusive in its non-sensuous indefiniteness,

appealed to Wordsworth as offering a better approximation

than most other words to the stir of strange energies around

(and within) him. But (if we may apply to Wordsworth’s

words what Leavis said about his visionary moments),

‘though Wordsworth no doubt was right in feeling that he

had something to pursue, the critic here is in a different case!’7

The plight of the critic or the reader has been described by

Sir Maurice Bowra where he speaks of Wordsworth and

Shelley talking about visionary experience in darkly abstract

words. ‘We see that something of utmost import is afoot and

THE CASE OF WORDSWORTH

143

that the poet is transported outside his usual self, but we hardly know what has happened to him. The skies open, and he soars on the wings of inspiration to explore the infinite, but an impenetrable obscurity hides his goal from our eyes.'8

This is probably overstating the case against words like 'motion', for no personal confession (whether direct or metaphoric) of mystic experience was ever free from obscurity. But it is true that when confronted with the mystical condition—a dark or dazzling but nonetheless concrete condition—Wordsworth generally used the abstract noun. To describe so specific and vivid an experience as the dissolution of the mind-matter disjunction (an experience he shook off as a boy by grasping a wall or a tree), Wordsworth (in the Fenwick note to the 'Immortality Ode') calls in a term from philosophical theory—'idealism'. And in the Ode itself, as he gets closer to the central mystery, he employs four gerunds in a row ('questionings', 'fallings', 'vanishings', 'misgivings') and then tries out a vivid image from Hamlet ('a guilty thing surprised'), only to revert the next moment to 'affections' and 'recollections'. The vocabulary of the notional is used to render the experiential. Wordsworth had taken over this vocabulary ('motion', 'presence', 'scene', 'fair', 'high', etc.) from the philosophy and poetry of the previous century. The nouns and adjectives of the eighteenth century generalize instead of discriminating, classify instead of specifying. As Donald Davie says, they 'turn their back upon sense-experience and appeal beyond it, logically, to known truths deduced from it'.9 They are the instruments of thought of an age when 'nearly everybody thought that He (God) could be deduced by tracing laws and classes, not perceived in a leap of insight'.10 Wordsworth, on the other hand, uses such words for rendering just such a leap of insight.

10

Suggestors of Emotion

The reaction against Bradley, inaugurated by L. C. Knights’s broadside, ‘How Many Children Had Lady Macbeth?’, in 1933, seems to have run its course. We notice nowadays that the character-monger and the theme-pedlar are elaborately polite about each other’s trades, and it is generally agreed that while characters are a part of the total design of the play, the design is not validated unless the characters are convincing ‘persons’. On the one hand, Knights, whose powerful ridicule had almost stilled all talk about character, is now heard saying: ‘ . . . And I for one would rather see among my pupils an honest and first-hand appreciation of what is offered by way of “character” than a merely mechanical working out of recurrent imagery and symbolic situations.’1 And for their part, the fine character studies and examinations of the character principle that we have had in recent years from John Bayley, John Harvey and others show that we have now a new generation of intelligent Bradleyans whose methods are much less vulnerable to the attack on character.

But not always. In Milton’s God, Empson argues that Milton’s Delilah is ‘a high-minded great lady’ and Milton’s Samson an ‘unintellectual’ nihilist and braggart. I will say nothing here about the techniques of character analysis which led him to this conclusion; nor ask whether the conclusion is congruous with a poem which is a very

144

SUGGESTORS OF EMOTION

145

conscious exercise in 'Tragedy, as it was antiently compos'd' and which is deeply personal, embodying (as Empson admits) a direct parallel to Milton's own life—because neither of these is my point here. My point is that Empson often gets away from the poem and into the Bible or into life.

Samson's statement to Harapha that his having married into the Philistine tribe the first time is proof of his goodwill is, Empson points out, disingenuous, since earlier on Samson had described the marriage as prompted by the Lord to enable him to find a casus belli. Empson's comment is:' . . . One might indeed argue that he is telling the truth now and was lying before, but the question is settled by Judges xiv. 4.'2 But is it? Can anything in the Bible, can anything outside Samson Agonistes, settle whether the Samson of Milton's poem is mendacious here or not?

The second example concerns Delilah. Empson obviously likes her. His comment on her visit to Samson is: 'It would be wilful to doubt that she still loves him and wants to help him, because we are given no other reason for her visit.'3 But we are given another reason—by Empson himself and in the very next sentence: 'She might indeed have a general political intention, to try to heal even now the divisions threatening civil war . . .'4 Now where did Empson get that? From the poem? No—for he says later: 'And I grant, of course, that she does not express the aim of reconciling the two parties; that would exasperate Samson and at best sound a useless excuse to other people.'5 When the text contains no evidence to support his guess, can the critic run up and fetch a reason from life or from the twilight realm of conditionals to explain away the absence of evidence? Is this Empson's alternative to the spatial-metaphorical

146

SUGGESTORS OF EMOTION

approach and its 'inhumanity and wrong-headedness', as he once called it?6 And if she had succeeded in her mission of love and peace and lived happily ever after with Samson, how many children would Lady Delilah have had?

No one, of course, would care to suggest today that what we know about a character should be no more than what the text explicitly tells us. But there is surely a difference between practising legitimate inference about a character and pursuing him down by-ways of speculation which lead away from the text. What the spell of the how-many-children game does to us is to make us forget that character is a function within the artefact's total being-what Knights today calls a 'vision of life—more or less complex and inclusive—whose meaning is nothing less than the play as a whole'7 and formerly used to call (employing terms more like Richards's and less like Wilson Knight's) the 'system of values that gives emotional coherence to the play'.8 Neither description would have surprised Sanskrit critics. They held, as we saw, that a poem is dominated by a cohesive principal emotion to which all the elements—plot, character, language—are more or less firmly oriented. (The dominant emotion is, of course, often amplified and diversified by the individual emotional motifs of the different scenes or episodes, and then a character may be aligned to the effect of the part, rather than the whole, of the play.)

Characters thus are to be valued less for their individuality and verisimilitude than for their relevance to, and enrichment of, the total emotional pattern. This anti-naturalistic position offered no problem to a poetic developed partly from classical Sanskrit drama which was 'non-realistic, conditioned by conventions that helped to govern the total response obtained by means of the language of each play'.9

SUGGESTORS OF EMOTION

147

These are actually the words in which Knights describes Elizabethan drama. Which explains why his notion of the relation of character to the play's emotional nexus—a notion developed from Shakespeare's plays—is so similar to that of the Sanskrit critics. The distinction between the actual (laukika) and the aesthetic non-actual (alaukika), between homo sapiens and homo fictus, and the status of homo fictus as no more than a factor of the total emotional impression are true of all writing, but they are more obviously true of writing that employs conventions and symbols than of writing that doesn't. The debate about character in Shakespeare continues, I think, largely because we cannot agree on whether Shakespeare is naturalistic, or non-naturalistic, or both by turns. But the idea that character is an element in the pattern of the play as a whole and can have no independence or absoluteness is one that applies essentially to all drama and fiction, all along the scale from realism (such as Arnold Bennett's) to its opposite which we can take T. S. Eliot, among many others, as representing. It is significant that both of them agree about character—at least in theory. Arnold Bennett noted in his diary after a meeting with Eliot: 'I was thinking about what T. S. Eliot and I had said about character in fiction . . . It must somehow form part of the pattern, or lay the design of the book. Hence it must be conventionalized.'10

We found that Eliot's essay on Hamlet studies the Queen not as a human or quasi-human being but as the objective correlative—an inadequate one—of the essential emotion of the play and that Sanskrit critics would have viewed her in much the same way. In studying the balcony scene in Romeo and Juliet, they would have described Juliet's status and function there as similar to the moonlight's, both being

148

SUGGESTORS OF EMOTION

the objective correlatives of love. This, we may feel, is pretty drastic. But if Juliet is the heroine (nāyikā) of the play, she is also, more importantly in the scene, the ālambanavibhāva of love, one among the many vibhāvas of love concentrated in the scene. The only effective insurance against the how-many-children error is a firm hold on the essential status of character as an element in the play’s emotional pattern. The naturalistic fallacy—the assigning of absoluteness and reality to characters—is, as we have seen, tenacious. And the Rasa-dhvani theory, precisely because it states so naïvely and coarsely that character, plot, imagery, rhythm and other elements have no raison d’être except as correlatives of emotion, can be uniquely serviceable as a prophylactic.

There are so many perceptive interpretations of The Ancient Mariner available now that it is tempting to get right back to Coleridge’s own well-known account of what he meant the poem to be. This is perhaps going outside the evidence of the poem itself and attending to a mere spelling out of conscious intention, but it might all the same be a useful exercise:

... the incidents and agents were to be, in part at least, supernatural; and the excellence aimed at was to consist in the interesting of the affections by the dramatic truth of such emotions as would naturally accompany such situations, supposing them real . . . it was agreed that my endeavours should be directed to persons and characters supernatural, or at least romantic; yet so as to transfer from our inward nature a human interest and a semblance of truth sufficient to procure for these shadows of imagination that willing suspension of disbelief for the moment, which constitutes poetic faith . . . With this view I wrote the Ancient Mariner . . .

SUGGESTORS OF EMOTION

149

We can, if we so choose, take the poet's statement at its

face value and view the poem simply as a presentation of

the marvellous. Walter Pater did so. The Sanskrit critics

would have done so; they would have had no difficulty in

identifying the ruling mood or emotion of The Ancient

Mariner as adbhuta (wonder). To them adbhuta was linked

with superhuman exploits (hence its special compatibility

with the heroic emotion) and, what is more relevant to

Coleridge's poem, with the intrusion of the supernatural.

And curiously enough, if the Romantic movement of

which The Ancient Mariner was an early flower has been

called the Renascence of Wonder, Viśvanātha and other

Sanskrit critics, speaking from classicist premises, also regard

wonder as underlying all response to literature.

There is an important difference though: in that while in

Sanskrit, the dominant emotion or mood of a poem or play

is the mood that rules the central character, this can hardly

be said to be the case in The Ancient Mariner. Neither the

central figure nor any of his associates or final rescuers

(except perhaps for the incomparable Pilot's boy) have

adequately evident emotional responsiveness attributed to

them. In the Biographia Literaria passage I have quoted,

Coleridge talks about his formula for establishing credibility:

which is, to interpolate in the poem realistic human emotion-

al reactions. This is precisely what does not get done in the

poem; the formula seems to have met with the same fate as

Wordsworth's principle of reproducing the real language of

men. Maybe the pace and dryness of ballad-style narration

ruled it out; maybe inset naturalistic behaviour would have

been incongruous. The experiences presented achieve credi-

bility not by any infusion of naturalism but in the only way

experiences in poetry can achieve credibility: by being made

150

SUGGESTORS OF EMOTION

to happen within the language of the poem. Our assent, for instance, to the preternatural sights that greet the Mariner on awaking from his sleep is obtained not through the presentation of the beholder’s sense of wonder but through the movement of the verse. ‘The hundred fire-flags sheen’,

To and fro they were hurried about!

And to and fro, and in and out,

The wan stars danced between.

In the wind that was heard roaring but did not blow on the ship,

. . . the sails did sigh like sedge.

And the continuous cascading of the lightning is no more than a function of rhythm in the lines that describe it. The truth of these marvels is guaranteed for us by the language and not by the human witness’s confession of wonder. The Mariner’s report on these is utterly reticent about their impact on his mind, so that it is impossible to see him as the vessel of the poem’s dominant emotion of wonder. This is true of the other characters as well, except, to be sure, for the Wedding-Guest, but he is outside the narrative.

Characters then are not objective correlatives of emotion in The Ancient Mariner. The burden of evoking the ‘particular emotion’ of the poem (to use Eliot’s terms again) is shifted from the ‘set of objects’ (whether men or spirits) to the ‘chain of events’. The whole plot is organized round a unified final emotional impression, which is wonder. It is possible no doubt to study the story in other terms; for instance, as presenting a moral vision. This can indeed be treated as one level on which the story and imagery operate, although there are serious difficulties here, the chief of which

is a structural one. With the return of love, prayer, sleep and rain and the falling off of the guilt badge, the sin-

penance-redemption cycle is nearly complete before the first half of the poem is over, and the second half, on this

showing, would be a gratuitous elongation of the final phase of the cycle. A moral concern there certainly is in the

story, but to assume that the overt affirmations of it in the commentary or in the poem itself and the less overt intima-

tions of the Christian imagery exhaust the significance of the poem or even that they establish the moral significance as the

central one is to disturb the proportioning in the poem.

When looked at instead as the suggestor of the poem's essential emotion, the plot in its second half will be seen not

as dragging tediously on but as justifying itself by its enrichment of the sense of wonder through a further chain of

situations and images: the cloud-burst under a bright moon, with the stars dancing between the fire-flags sheen and the

unbroken flow of lightning; the inspirited corpses, with the Mariner's knee touching his dead nephew's as the two of

them pulled at the same rope; the seraph choir; the crimson shadows in the bay and the seraph forms on the deck; the

sails of the returning ship as they appear to the hermit.

The second half of the plot thus continues and completes the objectification of the emotion of wonder. If valuing a

poem merely or mainly for its embodiment of an emotion is too hedonistic an approach, it is at least justified by the results

—viewed in these terms, the plot, as we found just now, moves easily into focus; and several logically inconsequential

situations, which used to be explained away as dream work, are identified as material contributory to wonder. To read

the poem thus is not to be like Goldsmith's school children ('And still they gaz'd and still the wonder grew'), for this

152

SUGGESTORS OF EMOTION

approach forces us to attend to the careful way the plot

is so structured that the principal emotion as the directing

force is maintained and yet (lest emotional unity should be-

come the fourth Unity and a blight) is diversified and

reinforced by means of ancillary emotions.

A modern poet would have risked touches of banality or

even comedy in the poem with a view to that great anti-

septic: irony. Coleridge, however, excluded anything that

would be in conflict with wonder (except, as we will see,

at the end), so that the wedding scene is strictly another

world momentarily glimpsed and even the Mariner’s re-

entry into solid familiar reality takes place in a hushed moon-

lit bay that looks more like dream than fact. So exclusively

does the plot devote itself to the one emotion that the poem

comes pretty close to being a monolithic rendering of it,

the result, remarkably, being in this case not monotony

but a rare concentration and tautness. A long poem needs a

unifying emotion more than a short one does, but implicit

in its length is the ineluctable fact that the emotion must ebb

when it has flowed and that its diminution will have to be

compensated. But in The Ancient Mariner, wonder never

eases its pressure, so that the poem has an unremitting in-

tensity characteristic of a lyric. It is like the preternatural

lightning it celebrates:

Like waters shot from some high crag,

The lightning fell with never a jag,

A river steep and wide.

Not that the poem is insulated from all emotions other

than wonder. It does in fact have its moments of joy, lone-

liness, weariness, anguish, and, above all, fear. Fear is the

emotion that recurs oftenest and combines best with wonder.

SUGGESTORS OF EMOTION

153

The wild aspect of the Mariner; the chilling sense of being pursued; the vision of the phantom ship; the dead crew assembled on the moonlit deck and mutely gazing—these and other moments of horror in the poem make the sinister and the macabre act upon and accentuate the purely marvelous.

But the crucial diversification of the central emotion is at the end of the poem. The reaching out of the Mariner’s heart to ‘Old men, and babes, and loving friends And youths and maidens gay’, linking itself with a similar reaching out earlier to the ‘happy living things’ below the ship, is an emotional state that is basically alien to wonder. We might call it loving-kindness. We might call it, as McDougall did, the tender emotion. In Sanskrit, it approximates to vātsalya, the tenth sthāyin. A sense of the sweetness of common company or of the beauty of familiar creatures, offsetting the confrontation of the unusual and the weird—in this juxtaposition of contrapuntal emotions the design of the poem completes itself.

11

Notes on Suggested Meaning

EVALUATE IT, OR MERELY EXPLICATE?

In the well-known passage in Seven Types of Ambiguity where he analyses Nash’s line, ‘Brightness falls from the air’, Empson lists all the meanings it suggests to him: the setting sun or moon; Icarus; the ‘glittering turning things’ mounted on the roofs of sixteenth-century buildings; hawks, lightning or meteorites descending on their prey; a bright earth under an overcast sky; static electricity from the hair; bright motes in sunbeams falling and becoming dust.1 Elsewhere in the book Empson says: “Most of the ambiguities I have considered here seem to me beautiful”2; and ‘I should claim, then, that for those who find this book contains novelties, it will make poetry more beautiful.’3 But is this claim true of his explanation of Nash’s line? Can we be sure that all the meanings he has discovered are of aesthetic worth?

To be fair to Empson, he does concede that he has often omitted explicit aesthetic evaluation but adds that it always precedes or succeeds analysis. ‘You think the poem is worth the trouble before you choose to go into it carefully, and you know more about what it is worth when you have done so.’4 The same point is made by Wellek and Warren in their Theory of Literature: ‘And sometimes the distinction is made between the “elucidatory” and the “judicial” as alternative types of criticism. But though separation between the exegesis of meaning (Deutung) and the judgement of value

154

NOTES ON SUGGESTED MEANING

155

(Wertung) can certainly be made, it is rarely, in “literary criticism”, either practised or practicable . . . an essay which

appears to be purely exegetical must, by its very existence,

offer some minimal judgement of worth; and, if it is exegeti-

cal of a poem, a judgement of aesthetic worth . . . To spend

time and attention on a poet or poem is already a judgement

of value. '5

But saying, ‘Well, we wouldn't be analysing the poem,

would we, if we didn't think it worth it?’ is only half the

answer. The approval implied in the act of choosing a text

for analysis is no more than a preliminary one and may be

confirmed or withdrawn as exploration proceeds and contact

is made with levels and areas of meaning. (Moreover, the

thrills of exploration or the excitement of standing up for

your own interpretation against your neighbour's can make

you oblivious of the original question of the worth of the

text. In fact the worth of the text soon seems proportionate

to the number of meanings and patterns detected, so that

conceivably the James Bond stories in which—mercifully

facetiously—a wealth of archetypal imagery was recently

discovered can take their place beside The Ancient Mariner

and The Waste Land.) But distinct from an initial implied

judgement of the value of the text, there is—or should be—

another process of evaluation constantly at work during the

analysis. Each time a fresh possibility of meaning offers

itself, the critic should ask, ‘But is this one poetic?’ He should,

of course, also ask, ‘Does the context warrant it?’ (if not

‘Can the poet have intended that?’) and possibly a few other

questions too, but his paramount question would be, ‘Is it

aesthetically valuable?’ If that is loaded or vague, we might

frame the question the way Holloway does: ‘ . . . How much

better does it make the actual passage? How much more

156

NOTES ON SUGGESTED MEANING

shall we value that, enjoy it, feel it contributes notably to the play?'6

The influence and authority of the sciences have invested description with immense prestige and this, in literary studies, has been at the expense of evaluation. I began with Seven Types only because it is one of the earliest essays, and still the finest, in a mode—explication des textes—that has since acquired more and more vogue till today it has become the largest concern of criticism and scholarship. The larger the number of explicators and texts and the farther afield interpretation ranges, the less time and inclination it has for value judgements, so that the mere fact that a meaning can be, or has been, read into a text is now treated as conferring validity on it; whate'er is is right.

The temptation to fetch a meaning that does not (to use Empson's words) 'give the line its beauty' is the pitfall that awaits all verbal analysts, and this is something that Sanskrit critics of the ninth and subsequent centuries were very much alive to. To them, a further meaning assumed did not automatically qualify to be classed as 'suggested'. As we saw in Chapter 4, it is only when the exclusive importance of the further meaning is established that it can be treated as a case of 'suggestion'; and one of the factors of exclusive importance is superiority in aesthetic worth. If the additional meaning does not have greater beauty (cārutva) than the primary meaning, then it remains as merely additional meaning and the passage does not graduate to the rank of suggestive writing. It has second-class status and is graded as 'poetry of subordinate suggestion'. Although this sounds rather like quality control in industry, the approach has its merits. The distinction between what is beautiful and what is not was sharp in classical Sanskrit poetry as it indeed

NOTES ON SUGGESTED MEANING

157

was in traditional literature everywhere, but it has broken down in our century. As C. Day Lewis says, 'for the modern poet nothing is inherently unpoetic simply because for the modern man nothing is inherently poetic . . .'7 Precisely because that is so, the criterion of beauty, which Sanskrit poetics regarded as central to the exploration of unstated meaning, now becomes even more important in validating the meanings assumed by the explicator. And the principle of felt beauty can be an effective corrective to the increasing cerebrality in the experiencing of poetry which is reflected in the products of verbal analysis.

WHEN IS IT PARAPHRASABLE?

When is poetry paraphrasable and when is it not? Is poetry paraphrasable at all? The Symbolist doctrine is that poetry is, by definition, unparaphrasable (form and content being one), but critics subscribing to the doctrine are not above the temptation to paraphrase. The second half (beginning 'Who then devised the torment? Love.') of the 'Little Gidding' lyric can hardly be said to be in the discursive mode. Yet the lines have been reduced to a neat paraphrase by Mathiessen, one of Eliot's finest critics:

We can hardly face the fact that love is essentially not release but suffering; and that the intolerable burden of our desires—our Nessus Shirt—can be removed by nothing within our power, but solely through grace. All we have is the terms of our choice, the fire of our destructive lusts or the inscrutable terrible fire of divine Love.8

On the other hand, the Movement liked to define poetry as a kind of moral discourse and somewhat testily affirmed the

158

NOTES ON SUGGESTED MEANING

legitimacy of paraphrase. But not all the Movement poets were always paraphrasable. Take, for instance, Elizabeth Jennings’s lines on what men felt when Lazarus rose from the dead:

This man was dead, I say it again and again.

All of our sweating bodies moved towards him

And our minds moved too, hungry for finished faith.

He would not enter our world at once with words

That we might be tempted to twist or argue with:

Cold like a white root pressed in the bowels of earth

He looked, but also vulnerable—like birth.

Obviously, paraphrasability is not always dependable as the badge of a tribe. Which brings us back to the question: When is poetry paraphrasable and when is it not?

According to Abhinavagupta, where it is an idea/fact (vastu) or a trope (alamkāra) that has been suggested, it can be expressed discursively as well; on the other hand, sug-gested meaning, when it is emotion, does not admit of statement. This is identical with the familiar position in Western theory that concepts are reducible to discourse and that untranslatable image is the only language of emotions. As a given shade or nuance of emotion that is being expressed is uniquely determined by the group of sensuous correlates discursively presented, it cannot be dissociated from them and expressed by any other means.

Only one ship is seeking us, a black-

Sailed unfamiliar, towing at her back

A huge and birdless silence. In her wake

No waters breed or break.

(Philip Larkin, ‘Next, Please’)

NOTES ON SUGGESTED MEANING

159

The lines suggest that the only certainty for us is death. The

metaphor that suggests the idea is easily resolved into literal

statement—in fact, long before Larkin wrote them, the lines

had already been paraphrased several times over by the

moralizing poets of the eighteenth and earlier centuries

(‘Awaits alike the inevitable hour’, etc.). A poem suggesting

death not as an idea but as an emotive event would be very

different:

A widow bird sate mourning for her Love

Upon a wintry bough;

The frozen wind crept on above,

The freezing stream below.

There was no leaf upon the forest bare,

No flower upon the ground,

And little motion in the air

Except the mill-wheel’s sound.

(Shelley)

The grey bleakness of the sense of death is evoked here by

images of the external scene. The emotion, presented in these

terms, cannot be done into any other terms.

I am not yet born; provide me

With water to dandle me, grass to grow for me, trees to talk

to me, sky to sing to me, birds and a white light

in the back of my mind to guide me.

I am not yet born; forgive me

For the sins that in me the world shall commit, my words

when they speak me, my thoughts when they think me,

my treason engendered by traitors beyond me,

my life when they murder by means of my

hands, my death when they live me.

(Louis MacNeice, ‘Prayer Before Birth’)

160

NOTES ON SUGGESTED MEANING

The first verse, where the archetypal symbols of joie de vivre potently operate, is not paraphrasable. The second is. What it suggests is a concept rather than an emotion: the all-powerful determinism of society, and the individual's absolute involvement in its evil.

Of course, the distinction between concept and emotion should be handled cautiously. An idea or fact not touched with emotion is hard to come by in poetry, except in palpably descriptive or ratiocinative verse, and there it would invariably be stated, never suggested. Abhinavagupta, who emphasizes the difference between suggested emotion and all else suggested in point of paraphrasability, also emphasizes elsewhere that all suggestion of idea or trope terminates in the suggestion of emotion. In fact, the Larkin passage we looked at cannot be treated as presenting death as an arid concept—the idea of death certainly comes enveloped in feeling.

As we noticed in earlier chapters, when the suggested meaning is emotion, so instantaneously does it spring from the stated meaning (viz. from the objective correlatives) that the stages of the process—almost similar to the discharge of lightning—are imperceptible (asaṁlakṣyakrama). On the other hand, a fact or idea is, so to speak, more matter than energy, and suggesting it is a less rapid process whose steps are distinguishable (saṁlakṣyakrama). This speculative version of how meaning is formed can help to explain why what is suggested cannot be isolated and presented independently in the one case but can in the other. The theory cannot be claimed to be anything more than an insight and is offered here for what it is worth.

NOTES ON SUGGESTED MEANING

161

DEFINING IT, INSTEAD OF MUFFLING

Sometimes

on fogless days by the Pacific,

there is a cold hard light without break

that reveals merely what is—no more

and no less. That limiting candour,

that accuracy of the beaches,

is part of the ultimate richness.

('Flying above California')

The quality of a lucid Californian day that Thom Gunn celebrates in these lines is also the quality of some of his poetry—certainly of large parts of the volume which accomplished his arrival in the full sense. That 'limiting candour', that defining overtness (with this important difference, that it is no part of richness of meaning, is not only no part of it but is basically alien to it) is a virtue he consciously practised in The Sense of Movement (1957). The poem 'Human Condition', which I examined in Chapter 6 as a case of suggesting a meaning only to 'de-suggest' or unfold it the next moment, is from The Sense of Movement. Image and explicated concept packed two-in-one fashion seems to have been the formula for most poems in this volume. Gunn's work has since gained so much in variety, suppleness and freedom that the mode of The Sense of Movement cannot be claimed to be his only, or even his characteristic, mode. Nevertheless, it might be rewarding to study it as an experiment in reversing the more common method which consists in muffling the tenor so that the vehicle is released from particularity of reference and its multiple meaning is enabled to operate as uninhibitedly as the contextual frame would let it.

162

NOTES ON SUGGESTED MEANING

'Round and Round', which I cited as the opposite of 'Human Condition' in method, is from Gunn's first volume, Fighting Terms (1954) and is typical of it. As the title implies, the poems present love and other themes largely in terms of fighting—through myths, that is, drawn from war and violence. The technique throughout is to offer the myth and avoid stating its import. Most of the myths are developed in some detail, even giving the impression of a one-to-one correspondence—but this is illusory, as their reference is imprecise and interpretable, and the poems are no more like allegories than the early Eliot is like Dante. The rewritten Achilles, Helen and Lazarus legends do point to something beyond themselves, but we are not quite sure to what. The reference of 'The Court Revolt', 'The Right Possessor' and 'Looking Glass' is less difficult to divine, and that of 'The Beach Head' still less difficult, but none of these poems is easy in the sense that it has a built-in explanation. Precisely this absence of a key makes Fighting Terms uniquely satisfying. 'The Secret Sharer', for instance, is a more convincing variation on the alter ego theme than 'The Monster' in the volume, My Sad Captains (1961)—more convincing because more reticent. In 'The Monster', Gunn is at pains to explain that the other self is a concretion of a ponderous consistency in love, luxuriating in bitterness and despair. The explanation inhibits the image.

The method of Fighting Terms got exchanged for its opposite in the next volume, The Sense of Movement. 'On the Move', in a way the title-piece and certainly the poem most typical of the volume, is a construct made up of declared image-concept equations.

They scare a flight of birds across the field:

NOTES ON SUGGESTED MEANING

163

The concept that this suggests is gratuitously set forth in the next line:

Much that is natural, to the will must yield.

Another example:

In goggles, donned impersonality.

The boys on motorcycles are

The self-defined, astride the created will.

Similarly, in ‘Thoughts on Unpacking’, the ‘sagging shapes’, almost as soon as they are presented, are identified by the poet as traces of past tensions and errors which still cling to the lovers and imperil their relationship.

Weightless withered leaves rustling on the ground are the recurrent image in ‘Autumn Chapter in a Novel’. Left to itself, it would have been an effective suggestor of the anaemic sentiment that brings the boy’s languid mother and the tutor together. But Gunn insists on explicating the leaves as the lovers’ words, sapless and ineffectual. The result soon enough becomes apparent. The poem moves to its climactic last line,

And leaves thrust violently upon the pane

and precisely at the moment when the image must act with most energy, it is found to have lost much of its suggestive power. In fact, ‘leaves’ is now less an image than a nonce substitute term for ‘words’.

Or take ‘At the Back of the North Wind’:

... Other smells,

Horses, leather, manure, fresh sweat, and sweet Mortality, he found them on the North.

The organic substances named here are between them quite adequate pointers to the North Wind being a symbol for

164

NOTES ON SUGGESTED MEANING

mortal life. The explicit use of the word ‘mortality’ de-

presses the poem from the level of symbol to that of state-

ment.

Gunn often gives in to a passion for abstract nouns like

‘mortality’ that reminds you of Wordsworth. He can, at

times, use them with defiant ostentation—almost like a

perverse child who seems to say, ‘Look what I am doing!’

In ‘The Beaters’, he places them where no one can miss

them—as rhyme words, and they are complete with the

morphemes that signal their abstractness. Thus: limitation,

affectation, resignation, devastation; loneliness, gentleness;

perversity, identity, liberty, extremity. The formal structure

of the poem owes a great deal to these conceptual terms

which bind each stanza together and round off ten out of the

twenty-seven lines in the poem. The theme of ‘The Beaters’

—the inflicting and receiving of pain—is picked up again in

the poem ‘Innocence’ in My Sad Captains; ‘Innocence’ offers

the myth and leaves it at that. The Sense of Movement does

include poems in this mode. If, for instance, ‘The Inherited

Estate’ can be read as a gracious compliment to American

culture, it is because the two dominant images—the follies

and the young tree—are allowed to work unaided. This,

however, is a departure from the prevailing manner of the

volume, which is to define and explicate suggested meaning,

thus transferring it from the potential to the actual level and

ensuring (to borrow Gunn’s words from a different context)

that our perception

rests on the things,

and is aware of them only in

their precise definition, their fine

lack of even potential meanings.

(‘Waking in a Newly-built House’)

NOTES ON SUGGESTED MEANING

165

When Gunn's imagery is described as precise, what is meant is apparently not that it is sharp and vivid (in fact Gunn's poetry is conspicuously non-sensuous) but that each image has a reference that can be particularized and is in fact particularized in the poem in so many words. As to what made Gunn elect this mode for the best part of The Sense of Movement, one can only guess. Possibly, the increased philosophic commitment obliged him initially to point up the concept each time. Possibly, in the case of some poems, his allegiance to the Movement inclined him to explicitness. This is mere speculation; what is clear from his subsequent work is that he soon ceased to be a prisoner of the image-plus-spelled-out-tenor formula, although he returns to it at will.

To look at Fighting Terms and The Sense of Movement as we have done, solely in terms of whether the tenor is stated or omitted, helps to account for the curious divergence in the critical responses to Gunn's early work. Alan Brown-john, for instance, complains of the obscurity of Fighting Terms: ‘ . . . the themes and the nature of the poet's thinking were frequently muffled by the over-elaborate metaphysical conceits . . . Many of the poems in Fighting Terms are very similar: the manner and invention exciting, the import muffled and vague because of the means employed...’ John Press, on the other hand, admires the lucidity of the title-piece of The Sense of Movement: ‘Gunn develops a complex metaphysical argument through a series of images which are exact symbols for certain emotional states and intellectual concepts.’ Both are, of course, right. Only, Mr Brown-john's blame is in effect praise and Mr Press's praise blame. Muffle the import is precisely what poetry ought to be doing. And an exact symbol is at best a poor symbol.

166

NOTES ON SUGGESTED MEANING

DISOWN IT AND/OR DISPLAY

Robert Graves’s poem, ‘Turn of the Moon’, tells how it is the moon ‘as she turns’ that brings all rain:

But if one night she brings us, as she turns,

Soft, steady, even, copious rain

That harms no leaf nor flower, but gently falls

Hour after hour, sinking to the tap roots,

And the sodden earth exhales at dawn

A long sigh scented with pure gratitude,

Such rain—the first rain of our lives, it seems,

Neither foretold, cajoled, nor counted on—

such rain, adds suddenly the next line which is the last line of the poem,

Is woman giving as she loves.

This is essentially the same as Thom Gunn’s method in The Sense of Movement, but it works in an altogether different way. When the suggested meaning is dramatically revealed in the last line (the effect being recognition no less than surprise), the line proceeds to react on the body of the poem, flooding it with a new, a human, meaning. The effect thus is enrichment, because the tenor that Graves makes explicit is not a concept but the human term of the image.

A piquant variation of this method is to offer the suggested meaning with innocent-seeming openness, preacing it with a palpable announcement; as much as to say, ‘N.B. The suggested meaning is as follows’. Vernon Scannell tries this out in ‘Incendiary’:

NOTES ON SUGGESTED MEANING

167

That one small boy with a face like pallid cheese

And burnt-out little eyes could make a blaze

As brazen, fierce and huge, as red and gold

And zany yellow as the one that spoiled

Three thousand guineas worth of property

And crops at Godwin’s farm on Saturday

Is frightening, as fact and metaphor:

The facts having been set forth, the promised explanation of

what they symbolize is appended:

And frightening, too, that one small boy should set

The sky on fire and choke the stars, to heat

Such skinny limbs and such a little heart

Which would have been content with one warm kiss,

Had there been anyone to offer this.

George Barker’s ‘A Sparrow’s Feather’ is an experiment

in another, and more provocative, refinement of the ten-

naming technique. Offer a compulsively suggestive image,

and as you do so, solemnly warn all concerned that it means

no more than it says. Barker tells how a young sparrow came

and died in an empty birdcage which he had filled with

birds of glass and paper and tin bits. The poem ends:

So there, among its gods,

that moaned and whistled in a little wind,

flapping their paper anatomies like windmills,

wheeling and bowing dutifully to the

divine intervention of a child’s forefinger,

there, at rest and at peace among its monstrous

idols, the little bird died. And, for my part,

I hope the whole unimportant affair is

quickly forgotten. The analogies are too trite.

M

168

NOTES ON SUGGESTED MEANING

The first part of R. S. Thomas’s sonnet ‘From Home’ is an account of how when people return from the high pastures in Wales to the towns, the soil full of grass seeds is transferred from their boots to the cracks in the pavement and puts out grass in due course. The poem continues:

There is no meaning between these lines.

I am not thinking of mixture of race

Or earth’s abundance . . . And yet, and yet:

The perpetual current of Welsh feet

Watering these grasses, addressed in vain

To the late greenness of their hearts!

First disown, then display—here is a combination of the two modes that is an advance on either. A statement of suggested meaning is redeemed by a show of naïve frankness. It is more than redeemed, it acquires a sophistication that positively justifies it, when it is prefaced with a tongue-in-the-cheek denial that there is any suggested meaning whatever.

We had in earlier chapters made it look almost like a doctrine that what has been suggested cannot then be stated without damage to the poem. If this is true as a general principle, the virtuosity of some contemporary poetry does establish that it is only true as a general principle and that there can be departures that more than justify themselves by their charming disingenuousness and by their daring.

THE NOT-GIVEN

Omission must have originated as a taboo in primitive communities which feared that the naming of certain things or people would bring ill luck. From a superstitious practice it developed in due course into a philosophical concept. Early

NOTES ON SUGGESTED MEANING

169

Indian philosophy and logic, noting that the total meaning

of a sentence is more than the sum of the meanings of its

constituent words, assumed that since the extra meaning

was conveyed by the relations between the words and not

by the words themselves, it could be said to have been

conveyed by suppression as distinguished from expression.

Neither the truth of this theory nor whether this notion of

suppression in conveying a judgement is the same as the

notion of poetic meaning being suggested instead of being

expressed is of interest to us here. It is the latter notion by

itself—poetic suggestion by omission—that concerns our

enquiry, and we find it occurring in the ninth-to-eleventh-

century Sanskrit criticism that we have been constantly

referring to, where we have critics speaking of matter being

presented through concealment (samvṛttyābhihitam—Ānan-

dayardhana) and attaining beauty through concealment

(gopyamānatayā labdhasaundaryam—Abhinavagupta). 'In a

symbol,' as Carlyle said much later, 'there is concealment

and yet revelation: hence therefore, by Silence and by

Speech acting together, comes a double significance.' By

the end of the nineteenth century this was a familiar principle

in Western thinking. As we saw in an earlier chapter,

Whitman, writing in 1888, declared that he rounded and

finished little; eight years later, Santayana spoke of the

study and enjoyment of 'the suggestion of the not-given,

rather than the form, the harmony of the given'.11 Now if

this was, as he called it, 'the evident characteristic of modern

genius' in Santayana's time, it is more so in ours. Silence as a

language is an influential cult today, and a blank page, we

are assured, is the perfect poem.

To be sure, explication of the suggested meaning (as

we saw in previous chapters and further in this chapter) does

170

NOTES ON SUGGESTED MEANING

affect the quality of a poem. But if there is anything more injurious to a poem than thorough explication of the tenor, it is thorough suppression of it. Complete omission of the tenor involves not only leaving it unnamed but also providing no hints or pointers in its direction, with the result that all reference is drained off from the representational content of the poem which then lies inert, asking nothing, offering nothing. Partial omission has the opposite effect—it makes the poem come alive. It can take the form of fragmentary statement that leaves you restless; it can also take the form of suggested meaning that is half-revealed so that a compulsive need for apprehending the rest is generated and various competing possibilities of meaning knock on the door. The reader’s imagination is soon working full throttle. This cannot happen when none of the suggested meaning is revealed, nor when all of it is.

Take, for instance, John Holloway’s poem, ‘Beast of Burden’:

So ‘Let me help!’ you twitter: and you make Gingerly as if to . . . then, quick, you let me slump Off-balance on the ground. And, ‘Why doesn’t my back break?’ You complain, and you gawk at the great hang-dog hump That bouldered all the door-jambs as I groped Into your house. Why this grotesque back-lash (You ask) I sweat so much to get roped, Knotted onto me? Use a gully! Slash ! Well . . . would you really have me fling away My millstone, stoneself, selfbreak counterpart? Don’t they bury men whose crime is felo-de-se At those coloured lights? Drive a faggot through the heart? Better look twice, hadn’t you, at my pack, See what it is drowsing on my back.

NOTES ON SUGGESTED MEANING

171

Except for a hint or two, offered in the words ‘the great

hang-dog hump’ and ‘drowsing’, the burden on the back is

a sinister mystery screened from view. The minimal nature

of the representation ensures that no more than an incipient

image of what is carried on the back results. If the picture

were complete, its very roundedness and clarity would lower

its suggestive potency.

If the vehicle is thus partially presented, the tenor is

glimpsed very faintly indeed. The glimpse is afforded in the

line,

My millstone, stoneself, self break counterpart.

This says enough, if only just enough, to energize the

reader’s imagination and make it work out what is not said.

If the line were removed, the poem would be an exercise in

total suppression and inarticulate to the point of self-ex-

tinction. And if, on the other hand, the line were to be re-

written in the motorcycle = created-will manner, the

poem would extinguish itself in a different way.

Omission is the very condition of the reader-participation

aspect of suggestion. And never was this function better

described than by Dryden in ‘The Dedication of the

Georgics’ (1697):

I must confess the Criticks make it one of Virgil’s Beauties, that

having said what he thought convenient, he always left somewhat

for the imagination of his Readers to supply: That they might

gratific their fancies, by finding more, in what he had written,

than at first they cou’d; and think they had added to his thought,

when it was all there before-hand, and he only sav’d himself the

expence of words.

It is ironical that this almost definitive account of how

omission works in suggestive writing should be offered by

a master of non-suggestive poetry.

NOTES

1. SUGGESTION TODAY

1 W. Empson, Seven Types of Ambiguity (London 1930), pp. 63-4.

2 C. Brooks and R. P. Warren, Understanding Poetry (New York 1938/1966), pp. 379-80.

3 E. M. W. Tillyard, Poetry Direct and Oblique (London 1945/1959), p. 29.

4 W. Righter, Logic and Criticism (London 1963), pp. 87-116.

5 I. A. Richards, The Philosophy of Rhetoric (New York 1936/1950), p. 40.

6 W. Empson, op. cit., p. 1.

7 E. M. W. Tillyard, op. cit., p. 84.

8 W. K. Wimsatt Jr., The Verbal Icon (New York 1954), p. 109.

9 T. S. Eliot, Selected Essays (London 1932), p. 300.

10 R. P. Blackmur, Language as Gesture (London 1954), p. 6.

11 Ibid., p. 19.

12 Ibid., p. 16.

2. SUGGESTION: FROM POE TO THE PRESENT

1 The Oxford English Dictionary, under 'Suggest' 3.

2 See F. W. Bateson, English Poetry and English Language (New York 1961), p. 52.

3 D. Perkins, Wordsworth and the Poetry of Sincerity (Cambridge, Massachusetts/London 1964), p. 100.

4 See D. G. Hoffmann (ed.), American Poetry and Poetics (New York 1962), p. 254.

5 See D. G. Hoffmann (ed.), op. cit., p. 318.

6 J. Isaacs, The Background of Modern Poetry (London 1951), p. 19.

7 A. D. Hope, The Cave and the Spring (Adelaide 1965), p. 7.

8 A. C. Bradley, Oxford Lectures on Poetry (London 1909/1926), p. 26.

9 W. B. Yeats, Essays and Introductions (London 1961), p. 161.

172

NOTES, PAGES 27-51

10 H. Read, Phases of English Poetry (London 1928), p. 122.

11 L. Abercrombie, The Idea of Great Poetry (London 1925), p. 19.

12 L. Abercrombie, Principles of Literary Criticism (London 1932/Bombay 1958), p. 38.

13 T. S. Eliot, Selected Essays, pp. 300, 315.

14 F. W. Bateson, English Poetry, a Critical Introduction (London 1950/1966), p. 20.

15 W. K. Wimsatt Jr., The Verbal Icon, p. 146.

16 J. Bayley, The Romantic Survival (London 1957), p. 189.

17 O. Barfield, Poetic Diction (London 1928), p. 133.

18 W. J. Bate (ed.), Criticism: the Major Texts (New York 1952), p. 183.

19 D. Perkins, op. cit., p. 100.

3. SUGGESTION THROUGH THE OBJECTIVE CORRELATIVE

1 T. S. Eliot, On Poetry and Poets (London 1957), p. 152.

2 H. Levin, Contexts of Criticism (Cambridge/Massachusetts 1958), p. 259.

3 See F. Kermode, Romantic Image (London 1957), p. 150.

4 T. S. Eliot, op. cit., p. 26.

5 T. S. Eliot, Selected Essays, p. 51.

6 W. K. Wimsatt Jr., The Verbal Icon, p. 38.

7 T. S. Eliot, op. cit., pp. 144-6.

8 A. B. Keith, The Sanskrit Drama (London 1924), pp. 276-7.

9 W. McDougall, An Outline of Psychology (London 1923/1949), p. 324.

10 R. Skelton, The Poetic Pattern (London 1956), p. 7.

11 K. Smidt, Poetry and Belief in the Work of T. S. Eliot (London 1949), p. 110.

12 F. O. Matthiessen, The Achievement of T. S. Eliot (New York 1935), p. 63.

13 W. K. Wimsatt Jr., op. cit., p. 37.

14 G. Hough, Image and Experience (London 1960), p. 17.

15 R. Williams, Drama from Ibsen to Eliot (London 1952), p. 17.

16 I. A. Richards, Principles of Literary Criticism (London 1934), pp. 290, 293.

174

NOTES, PAGES 52-69

4. THE LAMP AND THE JAR:

STATED AND SUGGESTED MEANING

1 I. A. Richards, Speculative Instruments (London 1955), p. 41.

2 Ibid., p. 54.

3 D. Davie, Articulate Energy (London 1955), p. 21.

4 T. S. Eliot, The Use of Poetry and the Use of Criticism (London 1933),

p. 151.

5 F. W. Bateson, English Poetry, a Critical Introduction, p. 19.

6 J. Bayley, The Romantic Survival, p. 220.

7 T. S. Eliot, op. cit., pp. 152–3.

8 J. Bayley, op. cit., p. 217.

9 E. M. W. Tillyard, Poetry Direct and Oblique, pp. 11–14.

10 W. Empson, Seven Types of Ambiguity, p. 62.

11 L. Abercrombie, Principles of Literary Criticism, pp. 39–40.

12 T. S. Eliot, Selected Essays, p. 299.

13 Ibid., p. 145.

14 Ibid., p. 300.

15 F. W. Bateson, op. cit., p. 20.

16 J. Bayley, op. cit., p. 219.

17 F. W. Bateson, English Poetry and the English Language (London 1950),

pp. 42–3.

18 Quoted in R. Skelton, The Poetic Pattern, p. 49.

19 W. Empson, op. cit., p. 26.

20 C. Brooks and R. P. Warren, Understanding Poetry, p. 106.

21 W. B. Yeats, Essays and Introductions (London 1961), p. 160.

22 C. Brooks and R. P. Warren, op. cit., p. 341.

23 W. Y. Tindall, The Literary Symbol (Bloomington 1955), p. 19.

5. SUGGESTION THROUGH METAPHOR

1 C. Brooks, Modern Poetry and the Tradition (London 1948), p. 22.

2 C. Day Lewis, The Poetic Image (London 1947), p. 54.

3 A. Tate, The Man of Letters in the Modern World, Selected Essays:

1928–1955 (New York 1955), pp. 64–77.

4 I. A. Richards, Principles of Literary Criticism, p. 240.

5 W. K. Wimsatt Jr., The Verbal Icon, p. 127.

NOTES, PAGES 69–103

6 A. MacLeish, Poetry and Experience (London 1961), p. 65.

7 I. A. Richards, op. cit., p. 240.

8 F. W. Bateson, English Poetry, a Critical Introduction, pp. 44–52.

9 H. Read, Phases of English Poetry, p. 127.

10 C. Brooks and R. P. Warren, Understanding Poetry, pp. 321–2.

11 W. B. Stanford, Greek Metaphor: Studies in Theory and Practice (London 1936), p. 101.

12 I. A. Richards, The Philosophy of Rhetoric, p. 127.

13 W. K. Wimsatt Jr., op. cit., p. 127.

14 W. Empson, The Structure of Complex Words (London 1951), pp. 333–4, 341.

15 See also K. Kunjunni Raja, Indian Theories of Meaning (Madras 1963), p. 266.

16 A. Tate, op. cit., pp. 76–7.

17 C. Brooks and R. P. Warren, op. cit., p. 364.

18 C. Day Lewis, op. cit., pp. 93–5.

6. STATING AND SUGGESTING BY TURNS

1 E. M. W. Tillyard, Poetry Direct and Oblique, pp. 11–15.

2 D. Davie, Purity of Diction in English Verse (London 1952/1967).

3 C. Day Lewis, The Poetic Image, p. 94.

4 J. Bayley, The Romantic Survival, p. 124.

5 T. S. Eliot, On Poetry and Poets, p. 26.

6 Ibid., p. 32.

7 Ibid., p. 74.

8 Ibid., p. 32.

9 Ibid., p. 76.

7. STATEMENT POETRY

1 E. M. W. Tillyard, Poetry Direct and Oblique, p. 10.

2 W. K. Wimsatt Jr., The Verbal Icon, p. 237.

3 L. Abercrombie, The Idea of Great Poetry, p. 49.

4 D. Davie, Purity of Diction in English Verse, p. 68.

5 Susanne K. Langer, Feeling and Form (London 1952), p. 223.

6 J. Wain (ed.), Interpretations (London 1955), p. 204.

7 T. E. Hulme, Speculations (London 1924), pp. 134–5.

176

NOTES, PAGES 105–26

8 F. W. Bateson, article on ‘The Language of Poetry’ in The Times Literary Supplement of 27 July 1967.

9 V. de S. Pinto and W. Roberts (ed.), The Complete Poems of D. H. Lawrence, vol. I (London 1964), p. 27.

10 Ibid., p. 28.

11 Ibid., p. 28.

12 Ibid., p. 184.

13 Quoted from one of D. H. Lawrence’s letters in F. O. Matthiessen, The Achievement of T. S. Eliot, p. 89.

14 V. de S. Pinto and W. Roberts (ed.), op. cit., p. 27.

15 P. Larkin, review of John Betjeman’s Collected Poems in Listen (Spring 1959), pp. 14–22.

16 T. S. Eliot, On Poetry and Poets, p. 32.

17 Ibid., p. 74.

18 Quoted from an unpublished lecture (1933) in F. O. Matthiessen, op. cit., p. 90.

19 T. S. Eliot, On Poetry and Poets, p. 82.

20 Ibid., p. 74.

21 R. Williams, Drama from Ibsen to Eliot, p. 239.

22 T. S. Eliot, On Poetry and Poets, p. 77.

23 C. Brooks, Modern Poetry and the Tradition, p. 112.

24 Ibid., p. 116.

8. SUGGESTION AS A CLASSICAL METHOD

1 J. Bayley, The Romantic Survival, p. 192.

2 W. K. Wimsatt Jr., The Verbal Icon, p. 150.

3 F. W. Bateson, English Poetry, a Critical Introduction, p. 20; W. K. Wimsatt Jr., op. cit., p. 147; T. S. Eliot, Selected Essays, p. 326.

4 T. S. Eliot, Selected Essays, p. 326.

5 F. W. Bateson, English Poetry, pp. 20–21.

6 W. K. Wimsatt Jr., The Verbal Icon, p. 144.

7 G. S. Fraser, ‘Approaches to Lycidas’ in The Living Milton, ed. F. Kermode (London 1960), p. 50.

8 L. C. Knights, Explorations (London 1946), p. 16.

9 Y. Winters, In Defence of Reason (Denver 1937/1943), p. 363.

10 See F. W. Bateson’s article on ‘The Language of Poetry’ in The Times Literary Supplement of 27 July 1967, where he classes ‘breath’

NOTES, PAGES 128–47

177

among 'the much too emotionally suggestive words' that modern

poetry tends to avoid.

11 Reproduced in the New Statesman of 5 March 1965.

12 T. S. Eliot, On Poetry and Poets, p. 26.

13 W. K. Wimsatt Jr., op. cit., p. 37.

14 C. K. Stead, The New Poetic (London 1964), p. 130.

15 T. S. Eliot, Selected Essays, p. 21.

16 See Philip Le Brun's article on 'T. S. Eliot and Henri Bergson'

in Review of English Studies (August 1967), p. 285.

9. SUGGESTION OR STATEMENT?

THE CASE OF WORDSWORTH

1 P.E.N. New Poems 1960 (London 1960), p. 24.

2 D. Perkins, The Quest for Permanence (Cambridge, Mass. 1965), p. 75.

3 Ibid., p. 47.

4 A. MacLeish, Poetry and Experience, p. 170.

5 R. G. Collingwood, The Principles of Art (London 1938), p. 111.

6 M. Bodkin, Archetypal Patterns in Poetry (London 1934), pp. 30–6.

7 F. R. Leavis, Revaluation (London 1959), p. 174.

8 C. M. Bowra, The Romantic Imagination (London 1950), p. 277.

9 D. Davie, Purity of Diction in English Verse, p. 48.

10 Ibid., pp. 41–2.

10. SUGGESTORS OF EMOTION

1 L. C. Knights, Further Explorations (London 1965), p. 203.

2 W. Empson, Milton's God (London 1965), p. 216.

3 Ibid., p. 220.

4 Ibid., p. 220.

5 Ibid., p. 222.

6 W. Empson, article on 'Hunt the Symbol' in The Times Literary

Supplement, 23 April 1964, p. 339.

7 L. C. Knights, Further Explorations, p. 190.

8 L. C. Knights, Explorations, p. 47.

9 Ibid., p. 5.

10 Miriam Allott, Novelists on the Novel (London/New York 1959),

p. 290.

178

NOTES, PAGES 154-69

II. NOTES ON SUGGESTED MEANING

1 W. Empson, Seven Types of Ambiguity, p. 26.

2 Ibid., p. 235.

3 Ibid., p. 256.

4 Ibid., p. xiii.

5 R. Wellek and A. Warren, Theory of Literature (New York 1949),

p. 250.

6 J. Holloway, The Charted Mirror (London 1960), p. 208.

7 C. Day Lewis, The Poetic Image, p. 106.

8 F. O. Mathiessen, The Achievement of T. S. Eliot, p. 191.

9 A. Brownjohn in The London Magazine for March 1963, p. 46.

10 J. Press, Rule and Energy: Trends in British Poetry since the Second

World War (London 1963), p. 193.

11 G. Santayana, The Sense of Beauty (New York 1896), p. 174.

INDEX

Note. Separate page references to the full bibliographical citations given

in the Notes section (pp. 172-8) are not included when the author or editor

in question is named in the main text and indexed accordingly.

Abercrombie, Lascelles, 27, 56, 73,

98, 124

abhidhā, 70

abhidhāmūladhvani, 74

Abhinavagupta, 36, 39, 158, 160,

169

adbhuta, 37, 149

Aeschylus, 21

ālam̉banavibhāva, 34, 148

alam̉kāra, 43, 158

alaukika, 147

Allott, Miriam, 177 n 10 (Ch.

Allston, Washington, 31

Ānandavardhana, 14, 15, 34, 40,

41-2, 43, 47, 50, 169

anaucitya, 35

anubhāva, 37, 42

anupapatti, 70

anvitābhidhāna, 52

Aristotle, 11, 47, 50, 76

Arnold, Matthew, 88

asam̉lakṣyakrama, 46, 54, 160

Auden, W. H., 62-4, 80, 88-90

avivakṣitavācya, 74

Barfield, Owen, 29

Barker, George, 167

Bate, W. J., 29

Bateson, F. W., 11, 28, 31, 53, 58,

73-4, 105, 122, 176 nn 3 and 10

Baudelaire, Charles, 24-5

Bayley, John, 28, 53, 58, 92-4, 121,

144

Beardsley, Monroe C., 33, 44

Beer, Patricia, 134

Bennett, Arnold, 147

Bergson, Henri, 12, 102, 132

Betjeman, John, 99, 110-13, 115,

118

Bhagavad-Gītā, The, 10, 128

Bharata, 35, 38, 39, 43-4

Bhaṭṭanāyaka, 38

bhāva, 37

bhayānaka, 37, 47

bībhatsa, 37, 48-9

Bible, The, 145

Blackmur, R. P., 13

Blake, William, 54-5, 86, 100,

142

Bodkin, Maud, 142

Bowra, C. M., 111, 142

Bradley, A. C., 26-7, 144

Bremond, Abbé, 27

Brooks, Cleanth, 8, 11, 61-2, 67,

117, 175 nn 10 and 17

Brownjohn, Alan, 165

Buddha, The, 128

Burke, Kenneth, 12

Carlyle, Thomas, 169

cārutva, 156

Cassirer, Ernst, 132

Chatterji, Nimai, 128

179

180

INDEX

Chaucer, Geoffrey, 4, 7, 73, 97,

101-2

Chiaromonte, Nicola, 29

citrakāvya, 55, 86, 99

Clare, John, 102

Clarke, G. C., 140

Coleridge, S. T., 11, 17, 18-19, 21,

76, 130, 139, 148-53, 155

Collingwood, R. G., 139

Cowper, William, 97, 100, 101,

102

Crabbe, George, 102

Crane, Hart, 75

Dante, 21, 123, 162

Davie, Donald, 53, 86, 99, 143

Denham, Sir John, 101

dhvani, 14-15, 39, 52, 59

dhvanikāvya, 86, 99

Donne, John, 70-2

Dryden, John, 17-19, 27, 48, 97,

102, 171

Eliot, T. S., 5, 12, 27, 31-5, 41,

43-4, 46, 48, 50-1, 53, 57, 64, 65,

76, 80-1, 83, 90, 93-5, 102,

112-16, 121-33, 147, 150, 157,

162, 176 n 3

Emerson, R. W., 10, 23

Empson, W., 7, 9-10, 55-6, 60-1,

78, 144-6, 154, 156

Foakes, R. A., 140

Foss, Martin, 69, 77

Fraser, G. S., 101, 176 n 7

Freud, Sigmund, 77

Frost, Robert, 61, 92, 95-6,

116-19

Gascoyne, David, 64

Gershon, Karen, 136

Goethe, J. W. von, 18

Goldsmith, Oliver, 54-5, 86, 100,

101, 102, 151

Gourmont, Rémy de, 31

Gower, John, 99, 102

Graves, Robert, 166

gunībhūtavyaṅgya, 61, 90, 99

Gunn, Thom, 87-8, 90, 161-6

Harvey, John, 144

hāsa, 36

hāsya, 36

Hawthorne, Nathaniel, 22

Herrick, Robert, 58-9

Hoffmann, D. G. (ed.), 172 nn 4

and 5

Holloway, John, 155, 170-1

Homer, 48

Hope, A. D., 26

Hopkins, Gerard Manley, 46

Hough, Graham, 46

Hughes, Ted, 64, 118

Hulme, T. E., 12, 22, 33, 102, 122-3

Isaacs, J., 24

Jarrell, Randall, 103

Jennings, Elizabeth, 158

Johnson, Samuel, 102

jugupsā, 34, 37

Kālidāsa, 37, 41, 44, 126-9

Kames, Lord, 17-18

Karunā, 36, 40, 47-50

Keats, John, 21, 32

Keith, A. B., 40

Kermode, Frank, 67, 173 n 3

Knights, L. C., 123, 144, 146-7

krodha, 36

Kṣemendra, 35

lakṣaṇā, 70

lakṣaṇāmūladhvani, 74

lakṣyārtha, 75

Landon, Courtney, 80

Langer, Susanne, 53, 97-100, 132

Larkin, Philip, 110, 158-9, 160

laukika, 147

laukika-kāraṇa, 37

laukika-kārya, 37

Lawrence, D. H., 108-10

Leavis, F. R., 142

Le Brun, Philip, 132

Levin, H., 173 n 2

Lewis, Alun, 4

Lewis, C. Day, 67, 80-1, 91, 157

Li Po, 60

Lollata, 38

Lowell, Robert, 84, 103-7, 110

MacBeth, George, 119

MacLeish, Archibald, 43, 69, 138

MacNeice, Louis, 159

Mahābhārata, The, 40, 48

Mallarmé, S., 2, 16, 24-5, 28, 33, 97, 102

Mammaṭa, 35, 139

Marvell, Andrew, 27

Mather, Cotton, 18

Mathiessen, F. O., 157, 173 n 12 (Ch. 3), 176 nn 13 and 18

McDougall, William, 42, 153

Milton, John, 10, 49-50, 81, 84, 144-5

Monroe, Harriet, 75

Moore, Thomas, 19, 21, 24

Morris, William, 27

mukhyārthabādha, 70

Mukulabhaṭṭa, 73

Nash, Thomas, 60, 154

nāyikā, 148

nirūḍha-lakṣaṇā, 71

Osborne, John, 48

Oxford English Dictionary, The, 172

Palgrave, F. T., 102

Pater, Walter, 54, 149

Perkins, David, 17, 29, 135-6

Pinto, V. de S. (ed.), 176 n 9 (Ch. 7)

Poë, Edgar Allan, 6, 15, 18-20, 62, 94, 102

Pope, Alexander, 101, 102, 119

Pound, Ezra, 33, 60, 80, 87, 94

prayojana, 70

prayojanavatīlakṣaṇā, 70

Press, John, 165

Raja, K. Kunjunni, 175

Rāmāyaṇa, The, 38, 40

Ransom, John Crowe, 13

rasa, 34-42, 47, 49-50, 52

rasa-doṣa, 35, 139

rasa-dhvani, 34, 39, 42, 46, 48-9, 148

rasika, 36

rati, 36

raudra, 36, 48-9

Read, Herbert, 3, 27, 67, 74

Richards, I. A., 9, 51-2, 68-9, 76, 146

Righter, William, 8

Rimbaud, Arthur, 136, 138

Roberts, W. (ed.), 176

Rochester, John Wilmot, Earl of, 98

Roethke, Theodore, 59

rūḍhi, 70

sādhana, 38

sādhāraṇikaraṇa, 45

sādhya, 38

sahrdaya, 36, 50

182

INDEX

śama, 37

samlakṣyakrama, 160

Śaṅkuka, 38

śānta, 37, 40, 48-9

Santayana, G., 169

Scannell, Vernon, 166-7

Schelling, F. W., 18

Schiller, F., 18

Schlegel, A. W., 18

Schlegel, F., 18

Schopenhauer, A., 102

Shakespeare, William, 31, 34, 41,

48, 53, 57, 72, 80, 94-5, 98, 108,

143, 147

Shelley, P. B., 21, 32, 64, 74, 80,

88, 121, 124-7, 130, 140, 142, 159

Shirley, James

6, 8

Sitwell, Edith, 58

Skelton, Robin, 42, 174 n 18

Smidt, K., 173 n 11 (Ch. 3)

śoka, 36, 47, 49, 74

śṛṅgāra, 36-7

Stanford, W. B., 76

Stead, C. K., 131-2

Stern, Gustav, 78

sthāyin, 34, 36-7, 39, 41, 42, 47, 153

Sundarakāṇḍa, 38

svaśabdavācyatā, 139

Swinburne, A. C., 3, 7, 27, 100-1,

121-2

Tate, Allen, 12-13, 68, 79-80

Tennyson, Alfred Lord, 10, 21, 74,

88

Thomas, Dylan, 3, 53, 58, 80, 121,

136, 139

Thomas, R. S., 5, 168

Thomson, James (B.V.), 121

Tillyard, E. M. W., 8, 10-11, 54, 65,

86, 97

Timrod, Henry, 23

Tindall, W. Y., 64

Tuve, Rosemond, 67

uddīpanavibhāva, 129

Upaniṣads, The, 128

utsāha, 36, 49

Uttarakāṇḍa, 40

vācyārtha, 52, 55, 75

vācyārthāpekṣā, 58

Valéry, Paul, 24

Valmiki, 40

vāsanā, 36

vastu, 158

vātsalya, 153

vibhāva, 37, 42-3, 148

vīra, 36, 47-50

Virgil, 17, 48, 128, 171

vismaya, 37

Visvanatha, 35, 149

vivakṣitānyaparavācya, 74

vyabhicāriṇ, 40

vyaṅgyārtha, 39, 52

vyañjanā, 14, 39, 70

Wain, John, 100-1

Warren, Austin, 70 n, 154, 178

Warren, R. P., 61-2

Wellek, René, 70 n, 154, 178

Whitman, Walt, 10, 23, 28, 169

Williams, Raymond, 46, 115

Wilson Knight, G., 146

Wimsatt, W. K., Jr., 10, 28, 33, 44,

69, 97, 121-2, 176 n 3

Winters, Yvor, 12, 125

Wordsworth, William, 4, 8, 17,

29, 81-3, 88, 97, 121, 134-43,

149, 164

Wyzéwa, 25

Yeats, W.B., 27, 58, 89, 90-3, 99 n,

100