SOURCE: Robey, David. “Anglo-American New Criticism.” In Modern Literary Theory: A Comparative Introduction, Ann Jefferson and David Robey, pp. 65-83. Totowa, N.J.: Barnes & Noble Books, 1982.

[In the following essay, Robey offers an overview of New Criticism, including an explanation of the theories of I. A. Richards, as well as the development of New Criticism in England and the United States, focusing specifically on the study of poetry.]

The term ‘New Criticism’ is usually used for the literary theory and criticism that began with the work of I. A. Richards and T. S. Eliot before the war in England, and was continued by figures such as John Crowe Ransom, W. K. Wimsatt, Cleanth Brooks and Allen Tate in the United States during the forties, fifties and sixties. Although it developed quite independently of the Russian Formalist/Prague School structuralist theory discussed in the last two chapters, there are some fundamental affinities between the two movements. Both rejected positivistic literary scholarship and called for a renewed attention to literature as literature; both insisted on the differences between literature and other kinds of writing, and tried to define these differences in theoretical terms; both gave a central role in their definitions to ideas of structure and interrelatedness, and treated the literary text as an object essentially independent of its author and its historical context. Considering that the New Critics seem to have known nothing at all about the work of the Formalists and their successors (it is not mentioned at all, for instance, in Wimsatt and Brooks's Literary Criticism: A Short History of 1957), these affinities are really very striking. It is the differences between the two movements, however, that will be of more interest to us here.

The New Criticism almost certainly constitutes the English-speaking world's major contribution to literary theory, and as such it has exercised until recently a dominant influence on the teaching of literature in the United States and, to a lesser extent, in Britain. In the last few years, however, with the entry of European literary theory into British and American academic life, its prestige has definitely been on the wane. Yet there are good reasons for continuing to read the New Critics' work. At the very least they are interesting because they formulated a number of assumptions about literature and literary study that still play a significant part in the academic world today. More importantly, their work still has considerable validity as a theoretical alternative to Formalism and structuralism, an alternative which may seem a great deal closer to many readers' and critics' feelings about literature and life. For while they always emphasized the special qualities of literature, and in particular the idea that the meaning or effect of literature cannot be explained by a process of reduction to ordinary modes of expression, the New Critics also insisted on its connections with the ‘real’ world, and on the contribution it can make to coping with the problems of everyday human existence. In contrast to Formalism and structuralism, the New Criticism was empiricist and humanistic. I shall try to explain further the significance of this distinction in the pages that follow.

I. A. RICHARDS: VALUE

The New Criticism began, as I said, with the work of Richards and Eliot; more exactly one can say that it began with the publication in 1924 of Richards's Principles of Literary Criticism, a radical, polemical programme for the study of literature that has had an enormous impact on British criticism and scholarship in the past fifty years. This work, together with The Meaning of Meaning, written with C. K. Ogden immediately before it, and Science and Poetry and Practical Criticism, which first appeared in 1926 and 1929 respectively, was published early in Richards's long career (he was born in 1893, and died in the United States in 1979), when he was teaching English literature at Cambridge. I shall limit my discussion of Richards's views to these four books, both for the sake of clarity and simplicity, and because they were the most influential of his writings. It must be said, however, that it is something of an injustice to him to do this, since his ideas and interests subsequently developed in a number of important ways.

Like Russian Formalism, Richards's early work turns its back on positivistic scholarship, and calls for a criticism that deals directly with the distinctive properties of literature; where he differs from the Formalists, however, is in defining these properties in terms of human experience and human value. While the Formalists had treated them as objective features inherent in literature itself, Richards's emphasis was on the reader's response to literature and on the evaluation of this response. The central question of criticism is ‘What gives the experience of reading a certain poem its value? How is this experience better than another?’ (Richards 1967: 1). As a result he has relatively little to say about questions of literary form. What he is interested in above all is analyzing the process of reading, and formulating criteria by which to evaluate the experience that reading produces.

According to Richards, therefore, critics need two things that they did not habitually possess at the time he was writing, and usually do not possess now: a theory of communication and a theory of valuation. Without these criticism lacks intellectual rigour, and is unable to justify itself adequately in a world in which the personal and social utility of the arts is increasingly called into question. As its title indicates, Principles of Literary Criticism sets out to remedy this deficiency, by developing a systematic theoretical framework within which criticism can be properly conducted.

The main point on which this framework rests is the distinction, already made in The Meaning of Meaning, between two different functions of language. The symbolic or referential function uses words to talk about the objective world, to point to things, as Richards puts it. The emotive function uses words to evoke subjective feelings or attitudes, by means of the associations that words carry with them. The clearest example of referential language is scientific prose, in which the evocation of feeling is as far as possible eliminated, and the writer's and reader's attention focused entirely on things. The clearest example of emotive language is poetry, which is entirely concerned with the evocation of feelings or attitudes, and in which the writer's and reader's attention is not, or should not be, directed at any of the objective relationships between words and things. Poetry is pseudo-statement, Richards argued; when we read it properly ‘the question of belief or disbelief, in the intellectual sense, never arises …’ (Richards 1970: 277). He thus comes close to Formalist/structuralist theory in denying the referential function of poetry, but differs significantly from it in his identification of the emotive with the poetic use of language; for Jakobson, it will be remembered, the emotive or what Jakobson calls the conative and the poetic are quite distinct. It is worth noting, however, that Richards's concept of poetry is similar to that which we have encountered in the previous two chapters, in that poetry for him is simply shorthand for literature that has aesthetic value; his belief was that the value of literature as a whole lay entirely in its use of the emotive function of language.

From this identification of the poetic and the emotive it follows that Richards is far less inclined than the Formalists to stress the difference between poetry and ordinary discourse. Or more exactly, while he does stress the difference between poetry and referential language, he also stresses that the experience which poetry produces differs only in degree, not in kind, from other types of emotive experience. Too many theorists have deluded the public by talking about one sort or another of ‘Phantom Aesthetic State’; in reality value in literature is explainable in exactly the same terms as value in any other form of human activity. This proposition is more easily understood when one realizes that Richards's general theory of value is a materialist one. Man ‘is not in any sense primarily an intelligence; he is a system of interests’ (Richards 1926: 21). It follows that what is good is simply the ‘exercise of impulses and the satisfaction of their appetencies’ (1967: 44), an ‘appetency’ being defined as a conscious or unconscious desire in the broadest sense. Life being what it is, unfortunately, our desires inevitably conflict, especially since we are social beings as well as individuals (Richards invokes Bentham here), and thus have to consider the needs of others as well as our own. The problem of morality is therefore above all a problem of organization; we have to reconcile our conflicting impulses with one another, in such a way as to allow them collectively the greatest possible degree of satisfaction. ‘States of mind,’ Richards says (1967: 45), ‘are valuable in the degree in which they tend to reduce waste and frustration.’

The difference between ordinary emotive experience and poetic experience, or any other form of artistic experience, is simply that poetry and art in general carry the reconciliation of conflicting impulses to an exceptionally high level. In this respect Richards's theory constitutes a materialist rewriting of the famous theory of the imagination in Chapter XIV of Coleridge's Biographia Literaria (Coleridge 1817: II, 12-13). For Richards the moment of poetic creation is characterized by an exceptional sensitivity or ‘vigilance’ towards the entire range of impulses relevant to the situation that is being expressed, and by the organization of these impulses into an ordered and harmonious whole. The conflict which characterizes our everyday feelings usually causes part of them to be inhibited or suppressed; in contrast poetry and the other arts ‘spring from and perpetuate hours in the lives of exceptional people, when their control and command of experience is at its highest’ (Richards 1967: 22). They renovate and enhance our reactions to life by disrupting established habits of response, and creating in us a state of equilibrium of a kind that other sorts of experience can rarely achieve. This leads Richards to a position which seems close to the Formalist principle of defamiliarization, when he says, for instance (1970: 254), ‘Nearly all good poetry is disconcerting’. But in contrast with the Formalists his emphasis is all on experience, not on form, and on organization, not on difference. Unlike theirs, his position is an essentially humanist one, in that it stresses the relevance of art to life. The critic is ‘as closely occupied with the health of the mind as the doctor with the health of the body’ (Richards 1967: 25). Moreover, poetry can do for us what religion and philosophy no longer can, that is tell us ‘what to feel’ and ‘what to do’; poetry, Science and Poetry concludes (p. 82), is ‘capable of saving us’.

It is this view of the experience produced by literature that causes Richards to emphasize the role of the reader rather than the author or the text. The difference between the critic's activity and that of the good reader is not really all that great. The critic must study the text with the maximum degree of attention to all its parts (Richards suggests (1970: 317) that four poems may be too many for a week's reading), but his aim is to arrive at the ‘relevant mental condition’ (Richards 1970: 11) associated with it, and then to judge this mental condition according to the principles outlined above. It should be noted that in emphasizing the experience of reading Richards does not make the kind of distinctions between reader, author and text that have played an important part in a number of modern literary theories. That is to say, the ‘relevant mental condition’ which the critic/reader must recreate within himself is assumed also to be the mental condition of the author. The poem is defined, technically, as the ‘experience the right kind of reader has when he peruses the verses’ (Richards 1926: 10), and the right kind of reader is the one who manages to recreate in himself more or less completely the collection of impulses which the poet expressed in the poem or, more exactly, the ‘relevant experience of the poet when contemplating the completed composition’ (Richards 1967: 178).

In contrast to those modern theorists discussed elsewhere in this book, who have insisted that the literary text is in a significant sense independent of its author, and that the author's and reader's experiences are not and should not be the same, Richards treats the text simply as a transparent medium, a mere vehicle for conveying the experience of the author to the reader. He never doubts that it is possible or desirable for the critic to recreate in himself the mental condition of the author; he only recognizes that it is difficult. Practical Criticism deals with the obstacles blocking the reader's approach to this ‘mental condition’. Richards asked a sample audience in Cambridge to describe their responses to a set of thirteen poems supplied without titles or the authors' names. The book analyzes the different mistakes of interpretation and evaluation that Richards saw in these responses, and seeks to identify their causes. Some causes, he suggests, arise from the texts themselves: the difficulty of apprehending the full range of meaning, and the inevitable differences in readers' reactions to sound and imagery. The other causes lie in the readers alone: the habits of thought and feeling that they bring with them to the text, which distort or block their response to it. Richards emphasizes the low degree of critical competence which, as a result of these different factors, his highly educated audience in his opinion revealed. But the book is written in the faith that such errors in readers' responses can be corrected, and that when this is done their way will be open to the poet's mental condition, and therefore to the correct experience of the poem. Interpretation is not really a problem for Richards, but simply a matter of approaching the text with the right kind of attention.

This aspect of Richards's work is worth stressing, because it expresses a belief which is taken for granted by a great deal of literary scholarship and criticism, and which from a more modern point of view may well seem somewhat naive. But it needs to be said that it is not a belief that Richards himself takes for granted. Much of his early writing is concerned with the theory of communication as well as the theory of valuation, with the result that his view of the author-text-reader relationship is placed within an elaborate and sophisticated theoretical framework. The details of this framework need not concern us here, except in one important respect. In The Meaning of Meaning Richards's discussion of communication takes place under the heading of symbolism or the theory of signs. As we have seen in the previous chapter, the terms semiotics or semiology are now usually used for the general theory of signs, which in its European structuralist versions has played an enormously important role in modern literary theory. If we contrast Richards's symbolism with the semiotics of the European structuralists, we shall be able to clarify a fundamental difference of principle between his view of knowledge and experience and theirs. The difference turns on the fact that his view is empiricist, whereas theirs very definitely is not.

Like Saussure's Cours, The Meaning of Meaning starts from the proposition that there is an essential disjunction between language and reality, that it is a superstition to believe that ‘words are in some way parts of things’ (Ogden and Richards 1936: 14). Words, the authors argue, are really a barrier between us and the world; we cannot escape from the structure of our language, and the structure of our language and the structure of the world are far from being identical. From this common starting-point, however, Ogden and Richards move in a quite different direction from Saussure's. For Saussure the meaning of words does not depend in any way on their relationship with things; it is wholly determined by the arbitrary and conventional structure of language. Ogden and Richards, in contrast, stress that words are used to ‘point to’ things, and that their meaning does in the last analysis depend on the things they are used to point to, their referents; language may be different from reality, therefore, but it nonetheless reflects it. For this reason they specifically criticize Saussure for ‘neglecting entirely the things for which signs stand’ (p. 6). Their position is thus an empiricist one, in that it rests on the principle that knowledge is the product of experience. The structuralist position derived from Saussure is anti-empiricist, inasmuch as its emphasis on the arbitrary and conventional nature of language leads to the view that knowledge is not in any way the product of experience. This aspect of Richards's thought is not so important for understanding his literary theory, since, as we have seen, he associated poetry with the emotive, not the referential, use of language. But it illustrates a radical divergence between the Anglo-American tradition as a whole and a great deal of modern European literary theory. In particular it is a fundamental feature of the American New Critics' work, to which I will shortly turn.

Some features of Richards's theory may now seem rather out of date: his notion that poetic language is purely emotive, his materialistic conception of literary value, his view of the author-text-reader relationship. But other features have become established parts of the Anglo-American critical tradition, and have by no means necessarily been superseded. These are his empiricism and humanism, and what we can call his organicist insistence on close reading, on careful attention to every detail of a literary text, on the principle that the text, like a living organism, functions through the interaction of all its constituent parts. One other feature of Richards's work has had little impact in this country, but a much greater impact, as we shall see, in the United States: the demand that criticism be founded on theory. Until very recently the theoretical issues raised by literary study have been largely ignored in British academic teaching and research. Richards's influence in Britain is thus to be found more in the practice of literary study than in the theory, in the form of undiscussed assumptions rather than systematically developed ideas.

Most of all this influence is to be seen in the practice of close reading, which has become an established part of English literature courses in Britain. Richards's influence in this respect has been much reinforced by the work of his pupil William Empson, the author of Seven Types of Ambiguity, first published in 1930. This is not the place for a proper discussion of Empson's views, which like a great deal of British work are more concerned with critical method than with theory (he wrote (1950: 594) that ‘a critic ought to trust his own nose, like the hunting dog, and if he lets any kind of theory or principle distract him from that, he is not doing his work’). But his practical, analytical development of Richards's theory has been an important model for British criticism during the last few decades, and I will therefore conclude this section with a very brief indication of the difference between his and Richards's approach. In his early works Richards outlined in a general way the different textual factors contributing to the effect of poetry; he distinguished, for instance, between the sense, feeling, tone and intention of a text in Practical Criticism (pp. 180-3), and in Principles of Literary Criticism (pp. 103-12) he offered some discussion of rhythm and metre. But since his principal concern was with the psychological effect of poetry on the reader, he did not carry this sort of analysis very far. His interest was to ensure that poetry was read with the right kind of attention, not to analyze or explain the textual means by which its effect is achieved. Seven Types of Ambiguity, on the other hand, is entirely occupied with this sort of explanation, through the ‘verbal analysis’ of ambiguity in poetry, ambiguity being defined (in the second edition) as ‘any verbal nuance, however slight, which gives room for alternative reactions to the same piece of language’ (Empson 1965: 1). Empson's work thus comes close to the theories discussed in the last chapter, which approach the literary text through its linguistic form; it differs from most of these, however, in being less systematic and theoretical, and in particular in being less influenced by developments in modern linguistics.

AMERICAN NEW CRITICISM: THE TEXT IN ITSELF

The American New Critics remained true to the spirit of Richards's work by emphasizing the distinctive properties of literature or poetry, and by dealing with them in a way which was not only empiricist, humanistic and organicist, but also theoretical as well. Their conception of literature differed from his, however, on at least one fundamental point: they were much less interested in the experience of reading than in the objective features of the medium, the literary text itself; and they therefore spent much less time on evaluation than on description and analysis. In this respect, clearly, they were a great deal closer than Richards to the Russian Formalists and their successors. But this shift of emphasis was not due to any direct influence from the Formalists whom, as I have said, the New Critics did not know; it was due to the influence of a figure much closer to home, the poet T. S. Eliot. The point is frequently made, quite rightly, that the American New Criticism is a development of Richards's and Eliot's work fused together.

Eliot's writings on poetry do not contain a systematic theory, but on two matters of principle he adopted a stance strongly opposed to that of Richards: he refused to accept either that poetry consisted in the use of emotive language, or that it was simply a vehicle for communicating the author's experience to the reader. ‘Poetry,’ he argued (1920: 52-3), ‘is not a turning loose of emotion but an escape from emotion; it is not the expression of personality, but an escape from personality.’ Good poetry objectifies feeling, expressing it indirectly through the description of things—through the ‘objective correlative’ (Eliot, 1972: 145). Good poetry is also equally concerned with thought; Eliot admired particularly the English Metaphysical Poets who, he suggested (1972: 286), ‘incorporated their erudition into their sensibility’. On such grounds as these he argued that the experience of the author and that of the reader must necessarily be different, that ‘what a poem means is as much what it means to others as what it means to the author’ (Eliot 1955: 130). From these rather brief suggestions the New Critics built up a much more definite and systematic theory of the literary text, and of the relationship between text, author and reader.

The term ‘New Criticism’ seems to have come into circulation with the publication in 1941 of a book of that title by the American poet and critic John Crowe Ransom. Like many terms of the sort, it does not have a very precise definition. The British and American critics for whom the label is commonly used, however, can be distinguished by a common rejection of established modes of criticism and scholarship (Empson and Leavis are sometimes included among them), and in the case of the Americans by the fact that they proposed alternative methods of criticism supported by well-developed literary theories. I shall limit myself in the rest of this chapter to the most interesting and influential of these theories, those of Ransom and the other Southern critics Cleanth Brooks and Allen Tate, and those of W. K. Wimsatt and his collaborator the philosopher Monroe Beardsley; all of these writers were active mainly in the forties and fifties. More exactly, since these five writers differ not only from other New Critics but also among themselves, I shall for the sake of clarity and simplicity concentrate on two texts, Wimsatt's The Verbal Icon and Brooks's The Well-Wrought Urn, referring to the work of the others only insofar as it corroborates or develops what is said there.

One further point that is often made to explain the nature of the New Criticism is that it developed outside the ambit of the main university graduate schools, in small colleges mainly in the South. This is particularly evident in the work of Allen Tate, who like Ransom was a poet as well as a critic, and was certainly not a scholar of the traditional type. Tate was a fierce critic of the ‘cloistered historical scholarship of the graduate school’, with its positivistic assumption that the literary text ‘expresses its place and time, or the author's personality’ and nothing more (Tate 1959: 7 & 54). Criticism, he argued, should be opposed to this kind of scholarship, and concern itself with the specifically literary properties of texts. The same opposition of criticism and scholarship and the same demand that criticism should concern itself with specifically literary properties are the inspiration of two of the best-known theoretical products of the New Criticism, the essays on ‘The intentional fallacy’ and ‘The affective fallacy’ written jointly by Wimsatt and Beardsley and published in the Sewanee Review in 1946 and 1949 respectively (now in Wimsatt 1958: 1-39). Together these two essays are an attempt to construct the theoretical basis for an alternative to positivistic scholarship, an alternative that will deal with the specifically literary properties of texts, and deal with them with the same degree of objectivity and rigour as scholarship has traditionally claimed.

Wimsatt and Beardsley argue that a poem (short-hand, as usual, for a literary work of art) is, and therefore should be treated as, an object in the public domain, not the private creation of an individual. The author's experience and intentions at the time of writing are matters of purely historical interest, that do not—contrary to the ‘intentional fallacy’—in any way determine the meaning, effect or function of his creation. As far as the author's experience is concerned, what counts from the viewpoint of criticism is only what is embodied in the text, and that is wholly accessible to anyone with a knowledge of the language and culture to which the text belongs. As for the author's intentions, what counts is only whether or not he has succeeded in writing poetry, and that too can be discerned by reference to the text alone. Thus most of what passes for literary scholarship is excluded from the sphere of criticism: studies of authors' lives, of their immediate environment, of their ideas about writing and of the genesis of their works. The only history that the critic must master is the history of words; he must grasp the full historical meaning of the language used in the text, including all its associations, and of the names to which reference may be made, but only to the extent that their meaning is a matter of public record about the culture in which the text was produced.

Wimsatt and Beardsley's approach is thus not a-historical, but it severely restricts the role of history in literary study, relegating questions about ‘how the poem came to be’ to a different, and by implication inferior, branch of enquiry. In this respect their position is not far removed from Richards's, since he too ignored the external history of texts, on the assumption that the ‘relevant experience’ of the author is entirely accessible from his work alone. But in another respect they take issue directly with Richards, following the guidance of Eliot. The ‘affective fallacy’, they maintain, is the fallacy of arguing, as Richards does, that poetry consists in the emotive use of language, and that the primary consideration of the critic must therefore be the effect that the poem has on the reader. Wimsatt and Beardsley's view is that a poem is not just a vehicle for conveying feelings, but an independent object with distinctive features of its own. To study the effect of the object rather than the object itself is to put the cart before the horse, since the cause of the effect is to be found in the object, and besides the effects of literary objects vary notoriously from one reading and from one reader to another. It is therefore essential to distinguish between effect on the one hand, and meaning or ‘cognitive structure’ on the other. It is with meaning, as the publicly accessible and thus objective constituent of texts, that literary criticism must be concerned; effect, being both variable and private, is much better left outside the field of enquiry. It is on this basis that the New Critics argued that criticism was capable of the same degree of objectivity and rigour as traditional scholarship could achieve.

If we follow Wimsatt and Beardsley this far, we arrive at a position that has had a great deal of attraction for literary critics, that the object of criticism must be the literary text itself. Moreover along with much modern literary theory, the New Critics also held that literary texts are texts of a special kind, and that the task of criticism is to give an account of this special character. The theory developed for this purpose by Wimsatt and Brooks (and in a rather different way by Ransom and Tate) takes as its starting-point Richards's (and Coleridge's) principle of the reconciliation of opposites, though the view it arrives at differs from Richards's in a number of important respects. From their views on the intentional and affective fallacies (Brooks seems to have agreed entirely with Wimsatt and Beardsley about these) it follows that this reconciliation of opposites must be seen not as an event in the mind of the author or reader, but as an objective fact about the text's meaning or structure. As Brooks put it, the New Criticism ‘is concerned with the structure of the poem as poem’ (Brooks 1962: 108). By ‘structure’ Brooks meant the organization of meaning in the text, which in the case of good literature (‘poetry’) possessed a different character from that of ordinary discourse. The chief property of poetry is coherence, not of a logical kind, but consisting in the harmonization of conflicting meanings or attitudes; poetry is objectively characterized, Wimsatt suggested (1958: 236), by a ‘wholeness of meaning established through internally differentiated form, the reconciliation of diverse parts’. Coherence is thus associated with complexity; the meaning of the text is the product of the interaction of its parts, but this interaction depends as much on their difference from one another as on their similarity. For Brooks, Wimsatt and Beardsley complexity and coherence together constitute the key considerations in the analysis of literary texts.

Before illustrating the practical consequences of these principles, it will be useful to clarify further the American Critics' relationship with Richards on the one hand, and with the Russian Formalists and their successors on the other. To take the latter first, the New Critics clearly come very close to the Formalists and the Prague School in the importance they attach to the ideas of structure and interrelatedness (they even occasionally called themselves formalists), and in their insistence on the objective character of criticism and the distinction of the author and the reader from the text. But their notion of structure is a good deal narrower than that of the Prague School, which included all the different levels of the text and not just its meaning; and they were not much interested in the ideas of difference, defamiliarization or deviance to which the Formalists and their successors attached so much weight. What the New Critics emphasized was convergence within the text rather than deviation from an external standard. As a result they were far less interested in literary innovation than the Formalists tended to be.

A further important difference between the New Critics and the Formalists and structuralists emerges if we return to the question of the New Critics' relationship with Richards. This difference concerns their conception of meaning, a conception which is crucial to their insistence on the objective character of criticism. For the materialist Richards meaning was to be explained as the use of words either to point to things or to evoke feelings—in terms of behaviour, in the last analysis. If, as he argued, poetry could not be seen as pointing directly to things, then to explain it as relating purely to feelings seemed to be the only alternative. Wimsatt and Brooks, on the other hand, approached the question of meaning from a more mentalistic point of view, and for this reason were able to argue that poetry related to knowledge rather than emotions, while still accepting that its function was not, like that of ordinary discourse, to point directly to things. Meaning for them was not to be defined in terms of use or behaviour, but as a mental entity independent in an essential sense (as ‘intension’) of the material world, although in another sense (as ‘extension’) related to it. At the same time, we have seen, they stressed that meaning was not a purely subjective phenomenon; as Wimsatt put it (1958: 10 and 24) one can distinguish between what is public and what is private or idiosyncratic about a poem, or between its meaning, which is an objective fact about it, and its ‘import’, the reader's subjective response. The specific property of poetry consists in the organization of the public, objective meanings of words, that is of the concepts and associations which a culture as a whole, rather than an individual, attaches to them.

Although they do not seem to have been familiar with the work of Saussure, Wimsatt and Brooks here might appear to be approaching his principle that the meaning of words is purely conventional and arbitrary. But in fact they stopped well short of this principle, because they believed that the peculiar organization of meaning in poetry led back in the end to the ‘real’ world. According to Wimsatt, for instance (1958: 241), ‘Poetry is a complex kind of verbal construction in which the dimension of coherence is by various techniques of implication greatly enhanced and thus generates an extra dimension of correspondence to reality, the symbolic or analogical’. The interrelations of the meanings and associations of words within a text (‘the dimension of coherence’) is in the first place, one might say, a purely mental event, but its effect is to modify, enrich or enlarge the reader's experience of the realities to which these meanings and associations refer. Art, Wimsatt and Brooks wrote in their joint history of literary criticism (1957: 743), ‘ought to have the concreteness which comes from recognizing reality and including it’. They did not believe, therefore, as the structuralists do, that structure in literature is a wholly closed system, and that language is a prison-house that shuts us away from reality. Their position keeps much of Richards's humanism, and is in the end an empiricist one, midway between Saussure's and Richards's views of meaning, or, as some might nowadays feel, an uncomfortable mixture of the two.

To sum up what I have outlined so far, the view of Wimsatt and Brooks is that the essential property of poetry consists in the reconciliation or harmonization of opposites; that this takes the form of an objective organization of the objective meanings of words; and that although the same organization generally cannot be found in other kinds of discourse, it nonetheless contributes to our knowledge and experience of ourselves and of the world. What specific forms, one must ask, does this organization take? According to Wimsatt it commonly consists in the use of analogy, especially through metaphor. In a really good metaphor, according to Wimsatt (1958: 149), ‘two clearly and substantially named objects … are brought into such a context that they face each other with fullest relevance and illumination …’ This means that one brings to bear on the other the full range of mental associations that the culture attaches to it; it is important to note that the objective meaning of words includes, for the New Critics, not only their dictionary definition (sometimes called their ‘denotation’), but also their associations (or ‘connotations’). An example of Wimsatt's (1958: 147-8) is what he calls the metaphor, and many would call the simile, in the last line of this passage from Donne's ‘A Valediction: forbidding mourning’ (like Eliot, the New Critics were particularly attached to the Metaphysical Poets):

Our two soules therefore, which are one,
Though I must goe, endure not yet
A breach, but an expansion,
Like gold to ayery thinnesse beate.

The comparison between the lovers' separation and the hammering of gold into leaf-form brings together two terms which are clearly quite different and therefore might justifiably be described as opposites; and the conjunction of meanings thus established creates a series of connections (the relationship between the separated lovers is like gold leaf in that it is ethereal (‘ayery’), delicate, easily damaged, but at the same time precious, pure, bright, etc.), which when related to real experience possesses considerable illuminating force. This is the sort of procedure that Wimsatt had in mind when he said (1958: 149) that ‘poetry is that type of verbal structure where truth of reference or correspondence reaches a maximum degree of fusion with truth of coherence—or where external and internal relation are intimately mutual reflections’.

The New Critics did not believe, of course, that the value of poetry lay in isolated metaphors or similes; on the contrary, they insisted that in good literature all the components of the text contributed to the reconciliation of opposites that constitutes its poetic function. Clearly there may well be more than an element of exaggeration in this insistence, but it makes more sense if we accept their view that a great many features of literature that might not normally be recognized, at least at first sight, as terms of a comparison, nonetheless have a metaphorical or analogical function. For instance events or situations in narrative poetry or prose can act as metaphors for a state of mind; an obvious example would be descriptions of houses in Dickens or Balzac. A narrative theme can be a metaphorical image of a psychological process; Wimsatt suggests that the relationship between the knight and the lady in Keats's ‘La Belle Dame Sans Merci’ can be read as an expression of the ‘loss of self in the mysterious lure of beauty’ (Wimsatt 1958: 80-1). In fact all good poetry, according to Wimsatt, is characterized by an intricate network of analogical relationships created by what he calls the ‘iconic’ properties of the verbal medium. What does he mean by ‘iconic’?

It is noteworthy that Wimsatt, like Richards, connects his view of literature to a semiotic theory. In Wimsatt's case the theory is that of the American behaviourist C. W. Morris, whose work follows in the empirical tradition established by Ogden and Richards. Wimsatt's use of the term ‘iconic’ (and the title of The Verbal Icon) derives from Morris's distinction (1971: 37) between the ‘iconic’ and the ‘symbolic’ sign; the former is that which ‘characterizes … by exhibiting in itself the properties of an object’, the latter that which does not do so, but has instead a purely conventional relationship with the object that it designates. Pictures are obvious examples of iconic signs; individual words, on the other hand, are generally symbolic signs, except for a limited number of cases such as those of onomatopoeia. Following Morris (1939), however, Wimsatt argued that one of the distinguishing characteristics of poetry was to exploit the ‘iconic or directly imitative’ powers of language (Wimsatt 1958: 115). This might seem at first sight to express the naive view that onomatopoeia and other kinds of sound symbolism constitute essential features of poetry, but Wimsatt's conception of the iconic function of language embraces a great deal more than this. ‘Iconicity’ in poetry was to be found, he argued, in metrical structures, in various figures of speech such as antithesis, and in general in the way in which words are arranged in sequence. Thus a disordered sequence of clauses or sentences can act as an iconic representation of material or emotional disorder. More importantly, the different types of phonetic, metrical, syntactic and semantic parallelism in poetic language indicate iconically a connection of meaning between the terms involved, and thus point to or reinforce the analogies between disparate elements from which, in his opinion, so much of the effect of poetry derives.

A random example of this process can be seen in two lines from Wordsworth's ‘Intimations of Immortality’:

Whither is fled the visionary gleam?
Where is it now, the glory and the dream?

Here the verse structure, especially the rhyme, together with the syntactic parallelism and the inverted semantic parallelism (visionary gleam/glory … dream), creates an exchange of effect between the two lines, whereby each illuminates the other. Because of the force of analogy, the meaning of the couplet as a whole is greater than the sum of the separate meanings of its two halves. This sort of analysis is substantially similar to Jakobson's discussion of Poe's ‘Raven’ (Sebeok 1971: 371-2), which I referred to in the last chapter, and it may well be that the New Critics' influence lay behind Jakobson's arguments there. It should be said, however, that Wimsatt's treatment of the iconic properties of language is as far as the New Critics went in the direction of stylistic analysis as I have defined it; in general they were much more interested in meaning than in forms of expression, and would have undoubtedly had scarce sympathy for Jakobson's attempt to define the properties of poetry in purely linguistic terms.

Finally, for both Wimsatt and Brooks a defining characteristic of poetry was irony. Here too they follow Richards, who used the same term to characterize the ‘bringing in of the opposite, the complementary impulse’ (Richards 1967: 197), which he held to be characteristic of all great poetry. In Wimsatt's definition irony is a ‘cognitive principle which shades off through paradox into the general principle of metaphor’ (Wimsatt and Brooks 1957: 747); according to Brooks, it is the ‘most general term that we have for the kind of qualification which the various elements in a context receive from the context’ (Brooks 1949: 191). ‘Irony’ here is clearly being used in a very broad sense to include the types of analogical or metaphorical relationship that we have so far considered. It might therefore seem a rather odd choice as a label, but Wimsatt and Brooks's reason for using it is that they view these analogical or metaphorical relationships as producing effects essentially similar to those for which the word irony is more commonly used, and which they also see as an important part of poetry. In all these cases a kind of reconciliation of opposites can be seen.

It is not easy to find a general definition of the normal meaning of irony, but it usually stands for a process by which the content of a statement is qualified either by the reader's attribution of a contrary intention to the author, or by the reader's awareness of factors that are in conflict in one way or another with what is being said. In this sense irony is closely related to paradox, which involves the association of conflicting elements within the same statement. Thus the term irony is used in something approaching its usual acceptance when Brooks associates it with Yeats's appeal to the Greek sages in ‘Sailing to Byzantium’:

Consume my heart away; sick with desire
And fastened to a dying animal
It knows not what it is; and gather me
Into the artifice of eternity.

That Yeats should speak of the ‘artifice of eternity’ evidently undermines in a sense the appearance of passion and sincerity with which he invokes the Greek sages, and thus can be said to bring about a kind of ironic reconciliation between his aspiration to a life free from Nature, and his rational awareness of his human limitations (Brooks 1949: 173).

What is achieved by devices such as this is not a relationship of analogy, as in the preceding examples, but, in Brooks's words (1949: 189), a ‘unification of attitudes into a hierarchy subordinated to a total and governing attitude’. This is also what Brooks sees in Keats's poetry in his analysis of ‘Ode to a Grecian Urn’, though here the notion of paradox rather than irony is invoked (1949: 139-52). Brooks emphasizes the internal contradictions of the poem: the fact that the urn is silent, but also speaks; that it is motionless, yet the scenes depicted on it seem to be alive; that unheard melodies are sweeter than those which one hears, and so on. What these contradictions add up to is the paradoxical point that the ‘frozen moment of loveliness is more dynamic than the fluid world of reality only because it is frozen’, that art is more vivid than life only because it is not alive. It is this paradox, according to Brooks, that is the main point of the poem.

It is interesting to compare Brooks's analysis with Spitzer's treatment of the same poem (Spitzer 1962: 67-97). Brooks's may not necessarily be much better, but Spitzer's undoubtedly attributes a more prosaic kind of meaning to the text, which he sees as expressing an (unreconciled) opposition between the ‘archaeological message’ and the ‘aesthetic message’ of the urn. The contrast may therefore serve to illustrate one major merit of Brooks's criticism and of the New Criticism in general: their use of ideas such as irony may seem exaggerated and confusing, but it makes the important point that the meaning of poetry, though possibly analyzable, cannot be expressed properly in the form of a conventional prosaic statement. ‘Irony’ is a strategic term that indicates the capacity of poetry to resist or elude our attempts to reduce it to conventional modes of expression. This capacity is the main theme of Brooks's important essay, ‘The heresy of paraphrase’ (Brooks 1949: 176-96). At the very most, Brooks argues, all that the practice of paraphrase can do is offer a crudely approximative outline of the poetic content of a text. The error of so many critics and scholars is to write as if the paraphrasable elements in literature constituted its substance, whereas the value of literature is to be found not in propositions, but in relationships, and these relationships are not logical, but imaginative.

I hope that I have made the merits of the New Criticism sufficiently clear in the preceding pages. As to their possible limitations, one's view of these depends, as it does in the case of most literary theories, on one's own theoretical position. Stylistic analysts are likely to hold that the New Critics paid insufficient attention to the linguistic form of texts since, apart from Wimsatt's discussion of the ‘iconic’ properties of poetry, they were really only interested in certain kinds of structure of meaning, and made no use of the tools offered by modern linguistics for the analysis of poetic effect. Structuralist critics are bound to disagree with the New Critics' insistence on the referential function of literature, with their view of literature's connection with the ‘real’ world. Post-structuralist critics will deny that literature possesses the organic unity to which the New Critics attached so much weight. An objection that can more usefully be discussed here, however, is the general principle formulated by Wimsatt, Beardsley and Brooks, and to some extent shared by Ransom and Tate, that literary criticism can and should be both evaluative and objective. That is to say, although the New Critics were less interested in evaluation than Richards, and more interested in description and analysis, they nonetheless assumed that their principles constituted criteria for distinguishing good literature from bad, and that good literature had an important part to play in human affairs. Since this is an assumption taken for granted by a good deal of modern criticism, we should consider some of the objections that can be made against the New Critics' arguments in its favour. These arguments in its favour concern, as we have seen, both the meaning and the structure of texts, and difficulties arise in connection with both of these factors.

The thesis of ‘The intentional fallacy’, that the meaning of the words in a text is, and should be treated as, a matter of public knowledge, seems wholly unexceptionable as far as the dictionary-definition (the ‘denotation’) of words is concerned; but it seems much more problematic when one takes account of the broader associations that words carry with them (‘connotations’). The New Critics argued both that these broader associations were an essential part of the meaning of literature, and that a high degree of consensus existed, at least among experienced readers, as to which associations were attached to which words. Yet while it seems conceivable that a group of critics, given an adequate degree of historical or philological knowledge, might agree as to which associations are possible in the case of a given word, there is considerable room for disagreement, since every word possesses a very wide range of associations, as to which of these are relevant for the purposes of interpretation. Moreover the interpreter of a text has to make a decision on the relative importance of the different elements in a work, and he has to find a way of relating these elements to one another. Interpreting the meaning of a text is not just a matter of adding together cumulatively the individual meanings of the words of which it is composed. How can an objective solution to these problems be found?

Positivistic literary scholarship solves these problems by looking outside the text for information about the author's intentions in writing, but as we have seen, the New Critics rejected this solution. For them the answer to the question of interpretation, and at the same time to the question of evaluation, is provided by the concept of structure; good literature or poetry is distinguished from bad literature or non-literature by an objective structure of meaning, the balancing or reconciliation of opposing attitudes or terms. If we accept this, then we might have an objective principle for distinguishing between the relevant and irrelevant associations of words in a text, for ordering its parts according to their relative importance, and for connecting these parts with one another. But is such a structure of opposing factors really an objective property of poetry, and can we therefore as a matter of principle allow our interpretation and evaluation to be guided by it?

There is more than one reason why we might not. First, we can question whether the concept is an adequate one for distinguishing poetry from non-poetry. It is easy enough to identify opposing terms or attitudes in a text, but how does one decide whether or not they are balanced or reconciled? For Richards the reconciliation occurred in the reader's mind, and therefore could be identified introspectively. For the New Critics it was an objective structural feature of the text, but they offered no precise structural criteria for identifying it. In practice, like Richards, they had to rely on intuition rather than analysis, or at best on the consensus of experienced readers, a consensus which, the recent history of criticism seems to suggest, is not all that often achieved. Secondly, we may well doubt that the reconciliation of opposites is really a sound principle for defining poetry, whether it is truly applicable to all the texts one would wish to include under the heading of poetry or good literature. For like a great many theories, it seems much more easily applicable to some kinds of text than to others; one can see quite clearly its possible relevance to the sort of literature that the New Critics generally preferred to discuss, the lyric tradition from Shakespeare, roughly speaking, to Yeats (Wimsatt and Brooks described their movement (1957: 742) as ‘neo-classic’); but it is much less easy to see its relevance to the novel, or to much modern avant-garde writing. This last objection can be put in a form applicable to many of the theories discussed in this book: that it may be a mistake to try to define the essential features of literature in absolute and objective terms; that such definitions as one may devise are likely to be relevant only to certain kinds or certain features of literature, or merely constitute one among a number of possible ways of approaching the subject.

As far as the New Critics are concerned, the best conclusion might be, not that their view of poetry was necessarily wrong, but that what was wrong was the claims they made concerning it. One may well agree with their rejection of positivistic historical scholarship, and their insistence on a combination of interpretation and evaluation in the study of literary texts. One may even agree that the best key to interpretation and evaluation is their principle of the reconciliation of opposites. One should also accept, however, that interpretation and evaluation are not objective statements about literary works, but statements about the interaction between works and their readers, because critics cannot avoid imposing their own views and their own preferences on the texts with which they deal. This is not to say that the best alternative to positivistic scholarship is the exclusive study of the reader's response, such as Richards advocated in his early works. Much more promising avenues are opened up by those schools of thought that combine a description of the structure of texts with an account of the knowledge and attitudes that readers bring with them and of the processes to which they subject them: phenomenological theory, which studies the process by which readers create meaning in a text with much more attention to the text itself than Richards was inclined to allow—but for which, regrettably, there is no space in this book; and some versions of structuralist theory, which will be discussed in the following chapter.

Further Reading

Of the four books by I. A. Richards mentioned in the text Principles of Literary Criticism is the best introduction to his work. A fifth book, Coleridge on Imagination, is of particular interest in view of the extent to which Coleridge's theory of the imagination influenced Richards's thinking. On Richards see Schiller's I. A. Richards's Theory of Literature and the more recent and brief essay by Butler, ‘I. A. Richards and the fortunes of critical theory’. The celebration volume edited by Brower et al., I. A. Richards: Essays in His Honor contains useful information about Richards's early work in Cambridge, and some interesting tributes from other New Critics. The most important collections of T. S. Eliot's writings are the Selected Essays and On Poetry and Poets.

The most important writings by Wimsatt and Brooks are collected in The Verbal Icon and The Well Wrought Urn respectively. Five further essays by Brooks are forceful statements of the New Critics' position: ‘Literary criticism’, ‘The quick and the dead’, ‘The critic and his text’, ‘The formalist critics’, ‘Literary criticism: poet, poem and reader’. Tate's theoretical writings are collected in Collected Essays and Essays of Four Decades. Monroe Beardsley's Aesthetics translates the New Critics' views into more technical philosophical language and puts them in the framework of a general theory of art. Ransom's The New Criticism has interesting critical discussions of Richards and Eliot, but proposes a theory of literature considerably less sound than that of Wimsatt and Brooks, though he has strong affinities with them. An important critical discussion of the New Critics in general is Krieger's The New Apologists for Poetry.

Morris's semiotics is most easily accessible in his Foundation of the Theory of Signs and Signs, Language and Behavior. For phenomenological approaches to literature, see Poulet's ‘Phenomenology of reading’ and Iser's The Act of Reading. Hirsch's Validity in Interpretation also draws on phenomenology and offers a theoretical account of the interaction between reader and text, though its conclusions constitute a defence—an exceptionally intelligent and powerful one—of traditional literary scholarship.