1963
From the pamphlet issued by the MLA’s Committee on Research Activities, The Aims and Methods of Scholarship in Modern Languages and Literature, ed. James Thorpe, 2nd ed. (New York: Modern Language Association of America, 1970), 69–81. First appeared in first edition of 1963, 57–69. When a second edition was called for, Thorpe wrote to Frye asking whether he wanted to revise his piece. Frye responded that he felt it “does not call for very much in the way of revision. It seems to me that when it does go out of date, or if it is out of date now, it ought to be replaced with another essay by somebody else” (NFF, 1988, box 60, file 7). Apart from changes of accidentals reflecting current uses of hyphens, commas, and spelling, there was only one small change in wording in the second edition (see n. 8). The work was translated into Japanese in 1972.
The previous essays in this series have dealt with the essential techniques to be learned by scholars in the humanities.1 What is meant here by criticism is a further stage in the scholarly organization of literature. It is not necessary for a literary scholar to become a critic in this sense, but he must be conversant with one or more of the techniques dealt with in the previous part of this handbook before setting up as a critic. I have attempted criticism myself without knowing very much about some of them, and I have never found my deficiencies to be anything less than a constant handicap. Still, if one’s interest is really in criticism, it is perhaps well not to become too expert in them. They are so demanding that one may easily fall into the error of assuming that, if criticism is posterior to scholarship, it must be a final whole of which the other scholarly techniques are preliminary parts, and so be something requiring superhuman abilities.
Such psychological hazards may be more common in Classical scholarship, where literary criticism seems to be somewhat deprecated. A.E. Housman speaks in this exaggerated way of criticism at the opening of his essay The Name and Nature of Poetry, which, it is true, is not very successful as criticism, though for other reasons than his intense specialization as an editor of Classical texts.2 There is less diffidence in the criticism of modern literature: in Mr. Stuart Gilbert’s commentary on Joyce’s Ulysses, for instance, we are told that as Ulysses, like all great works of art, is a microcosm of the universe, a knowledge of the universe will provide a solution to the obscurities of Ulysses.3 This is closer to the kind of exuberance the critic needs for tackling difficult and complex literature. In the modern field, at least, a much greater hazard for critics than stage fright is the general fuzziness about the aims, purposes, limitations, and scope of academic criticism. And, as critics are occupationally disposed to write most about what puzzles them most, this uncertainty has already produced a bulky literature of its own, which has done more to express the confusion than to clear it up.
Much of what we are here calling criticism has traditionally been regarded as something different from scholarship, a matter of taste, feeling, subjective reaction, or sensitivity. Statements produced by such factors, like the statements of literature itself, are pseudo-statements in the sense that they cannot be verified or refuted. It seems therefore difficult, if not impossible, to establish standards that will distinguish good from mediocre criticism. Every scholar develops something analogous to a sense of smell, and if a book has declared itself intellectually bankrupt on page 2 the experienced scholar, left to himself, will not read on to page 502. But he may not have only himself to consider, and may have to give reasons for rejecting the book. One can perhaps get some lead from the style: if the writing is illiterate the handling of the subject is likely to be insensitive. A surgeon with rusty knives and dirty scalpels does not inspire confidence, nor does a critic whose style is rusty with abstraction and dirty with jargon. But this is only a lead and nothing more, as some badly written books do say something, and there are elegantly written ones which are as empty as Marlowe’s tomb, if less absurd than the conception of criticism that opened it.4
It is not socially acceptable merely to say in a review, “Mr. X has written a book on the subject Y, a subject his intelligence is not sufficient to grasp,” and if it were the chances of a reviewer’s being in a position to make such a remark are remote. Hence mediocre books are often treated with much the same respect as good ones, and get on all the bibliographies as part of the graduate student’s obstacle race. It is in fact a standard rhetorical device to begin an article with a footnote referring to the contributions of twenty-seven predecessors to the same subject, in which the three critics who know what they are talking about are lumped in with the twenty-four who do not, as part of the chaos out of which the new article is creating order. It is not much wonder, then, that many scholars should regard literary criticism proper as an inferior mental discipline to historical criticism or editing, where at least there are definite facts to get wrong.
In this essay I am dealing with scholarly or academic criticism, which I should connect with the other scholarly disciplines treated in this handbook, and which I should want to distinguish clearly, though not in the least pejoratively, from journalistic criticism or reviewing, much of which, of course, is done by the same people. There is a traditional metaphor which makes the critic the “judge” of literature. Such a metaphor may imply that Shakespeare and Milton and other impressive people are, relatively to the critic, in the role of prisoners or petitioners, a prospect so exhilarating that many critics wish to leap into a judicial role at once, on the Alice-in-Wonderland principle of sentence first, verdict afterward. I am aware of the weight and influence of critics today who insist that criticism is primarily evaluative, and my next sentence, whether right or wrong, has been carefully considered. The metaphor of the judge, and in fact the whole practice of judicial criticism, is entirely confined to reviewing, or surveying current literature or scholarship: all the metaphors transferred from it to academic criticism are misleading and all the practices derived from it are mistaken. The reviewer of a current book, whatever its content, is expected to lead up to a value judgment, to give a clear indication of whether or not he thinks the book worth reading. But an academic critic, concerned with the scholarly organization of literature, is never in this judicial position. He is dealing with a body of literature which has all, whatever its merits, been accepted as a valid subject of scholarly study. For current literature there is an audience that wants to select its reading, but for Lydgate’s poems or Heywood’s plays there is no such audience—except students, and selecting for them is a function of teaching and not of criticism. Yeats, writing an essay on Swift, remarks that he is unable to check a remark made by a critic about Swift’s letters to Vanessa, because those letters bore him.5 He is really saying that he does not wish to be a scholar, for no scholar can afford the luxury of being bored by anything that is in the least relevant to his area of study.
It is worth insisting on this point, because the whole concept of academic criticism, as something which grows out of and completes the work of scholarship, is a relatively new one. One feels that criticism has played a rather minor role in the literature of the past; that, especially in English literature, it has not had enough authority to affect the main body of literary creation, and that it is extremely fortunate for literature that it has not had such authority. The reason for this is that most criticism in the past, as well as much of it in the present, is judicial criticism, which is very apt to go wrong when it assumes an academic function.
Academic criticism is partly historical, studying past literature in its original context, and partly an attempt to express what past literature can communicate beyond its own time to ours. Some tension between these two attitudes is inevitable, and neither should gain a complete victory. Overstatements of historical criticism tend to belittle great writers by forcing them down into a historical process. Take statements of the type: “Classical writers are A, B, and C, whereas Romantic writers are D, E, and F” (the letters stand for contrasting groups of adjectives). In some contexts such statements may be useful, but if they lead to assuming that Dryden or Keats, by virtue of being labelled one or the other, was able to present only half a vision of reality, they clearly lose their usefulness. This is one extreme; reading everything in the past by our own standards, which means turning all criticism into the judicial process we use in reviewing a contemporary book, is the other. Rymer’s essay on Othello is the stock example of an extreme assimilation of academic to judicial criticism.6
We have to avoid of course the blunder that is called the intentional fallacy in criticism. The question, “What did the author mean by this?” is always illegitimate. First, we can never know; second, there is no reason to suppose that the author knew; third, the question confuses imaginative with discursive writing. The legitimate form of the question is, “What does the text say?” But what a text says it says in the language of its own time, and although that language can be intelligible to us, we have to try to read Milton’s theology and Swift’s politics precisely as we read Langland’s West Midland dialect, without translating. I say try to read, because reading something in the language of its own time is always, beyond a certain point, a conjectural reconstruction. If we ask what Samuel Beckett or J.D. Salinger meant to their own time, we find ourselves confronting a vast fog, and the original audience of Shakespeare or the original purchasers of Paradise Lost were no better off. Critics of the twenty-third century, if there are any, will insist, and rightly, that Beckett and Salinger should be related to the assumptions of their own age; but we do not know what those assumptions are, nor do Beckett or Salinger.
There has always been a view that the writer’s function is to instruct and delight. Judicial criticism, at least of literature proper, puts a strong emphasis on the latter: writers of plays, particularly, must please to live. But if we think of “instruction” not as illustrating moral precepts, but as expanding our vision, then instruction is quite as important in literature, and our instinct for regarding “escape” reading as to some extent sub-standard partly justified. Every writer, past or present, big or little, is, by the act of writing, making a bid for authority, for filling a place in our imaginative experience that no one else can fill in quite the same way. The first task of a responsible judicial critic dealing with a living and not yet established writer is to try to determine what the quality of his authority is. In academic criticism the authority has for most writers already been established. One may dislike Donne or Shelley, but it is critical malpractice to pretend that they are not important poets; writers of the past who have no importance in themselves still carry, for academic purposes, something of the authority of the literary traditions and conventions that they so pallidly, and therefore so lucidly, illustrate.
The academic critic needs, not judiciousness, but humility, though he needs that less as a moral virtue than as an intellectual precaution. What gives a judge the right to be on a bench is knowledge of law, and what gives a critic the right to be a critic is knowledge of literature, but there the parallel stops. If the critic is faced with a major work of literature, he is the one being judged, and the best he has to offer is still none too good. What is required from him is not evaluation, but, first, understanding as little inadequate as possible, and, second, some power of communicating that understanding which will make him worth reading as well as his subject. It is not difficult to give an impression of raising one’s standards by limiting one’s sympathies, but this is not a legitimate academic ambition.
The scholarly critic must never forget that his permanent value as a critic depends exclusively on what he accepts. T.S. Eliot is a critic of permanent value because of what he says about Dante, Donne, Dryden, Johnson, Marvell, and other poets he likes and appreciates. He does not need to write a second essay on them trying to correct what he said previously; nothing he has said need cause his reader the least embarrassment or uneasiness. When he writes of Milton and Shelley he is in the position of a bridge player saying “I pass”; we feel that we must wait for another critic who will do better with them. Yeats is a critic of permanent value because of what he says about the Irish National Theatre, about Blake, or about Shelley; when he writes of T.S. Eliot or Wilfred Owen he is merely demonstrating where his limitations are. In other words, most of a critic’s mistakes are rejections, and if he is dealing with the established dead, all his rejections are mistakes.
Even in reviewing, a “this will never do” critique is of limited usefulness, because if it is justified the review will seldom outlive its subject, and if it is not the best that can happen to the reviewer is a booby immortality.7 But in academic criticism the relevant value judgments are all implicitly assumed. One cannot work on a writer without implying that there is some reason that makes working on him worthwhile, and the normal reason is his intrinsic merit as a writer. When it is not, value judgments may point out the fact.8 Thus C.S. Lewis, in The Allegory of Love, devotes several pages to John Rolland’s Court of Venus, mainly for historical reasons, and therefore makes explicit critical judgments warning us that if we read the poem we may not find it very exciting. This is an example of value judgments playing a legitimate and essential role in academic criticism.9 But the academic critic’s function is to add to our understanding of the writer, and this is bound to make his merits, if he has any, more obvious. To praise or blame the writer, to inflate or deflate his reputation, is not the academic critic’s primary function at all.
In judicial criticism the standard on which the critic’s taste is founded is usually the actual or ideal taste of his own time, for which he is the spokesman. The unconventional is normally regarded with suspicion or hostility, unless it is conventional to be unconventional, though even that takes conventional forms, like the “angry,” “beat,” and “absurd” schools of writing in our day. The literature of the past conforms to the conventions of its own time, and hence is usually unconventional to the judicial critic as such. A Restoration or Victorian judicial critic of Shakespeare, or a modern critic like the late Wolcott Gibbs of the New Yorker, will tend to judge Shakespeare by the qualities that would please him if they were exhibited by a contemporary. The view of history implied is an evolutionary or progressive one. In some respects Shakespeare shows the correctness (Restoration), the refinement of manners (Victorian), or the liberality of social views (modern) that is required of a sensitive contemporary writer. To that extent he has something to say to us. If he shows negligence of the rules, coarse manners, or illiberal political and social views, the assumption is that he illustrates a cruder and more barbaric time, however this assumption may be modified or rationalized in statement. The history of critical judgments on Shakespeare, or any other major author, will show every variety of such attitudes, and no critic can avoid them if his criticism is fundamentally judicial. Edmund Wilson is perhaps the finest judicial critic of our time, with an academic orbit of about a century or so, and it is instructive to see how his judgment goes out of focus when dealing with anything outside that orbit, such as Swift or Ben Jonson or the Book of Genesis. The judicial critic who is only incidentally a critic, and primarily a poet or novelist himself, has a still narrower range, as he is likely to be, at least unconsciously, using his own work as the criterion for his judgments. The most notorious examples of critical duncery, such as the contemporary reviews of Keats’s Endymion, can still hardly be compared, in their lack of comprehension, to many of the things that great writers have said about one another.
Academic criticism is part of the systematic study of literature of which the other parts are the subjects dealt with earlier in this handbook. Its approach is categorical and descriptive: it tries to identify a writer’s work. Now even judicial criticism never leads logically to a value judgment: value judgments may be assumed at one end or emitted at the other, but the relation between them is rhetorical only. The source of the confusions involved here is the failure to distinguish criticism from the direct experience of literature. Direct experience, developed by practice and habit, is the basis of good taste, and the normal result of good taste is a value judgment. Good taste in itself is inarticulate: it feels and knows, but cannot speak. Value judgments may be asserted, intuited, assumed, argued about, explained, attacked, or defended: what they never can be is demonstrated. I may know that Wallace Stevens is a better poet than a writer of inspirational doggerel in a newspaper, but I can give no evidence to anyone who likes the doggerel, except my own opinion and that of others whom I respect because they agree with me. I can, of course, analyse a passage from Stevens and demonstrate a density of texture in it that the doggerel does not have, but that proves nothing unless I am willing to assert that poetry with this density is always better than poetry without it. Such an assertion, besides being wrong in itself, would still leave me unable to demonstrate the superiority of Sir Patrick Spens or Burns’s Farewell to Nancy, which I also prefer to the doggerel.
The same thing is a fortiori true of academic criticism. If a critic writes a book on Keats, his feeling that Keats is a superb poet may, and should, permeate all his writing. But his criticism will never prove Keats’s merit or value or superiority to any other poet; the fact that he has written the book is his testimony to that value, and such testimony implies only; it cannot infer or be inferred. It is on this basis that we have to determine the relative authority of judicial and academic criticism. Judicial criticism is based on good taste, and good taste is a skill founded by practice on the knowledge the critic has; academic criticism is a structure of knowledge. Lapses of taste and value judgment, when made by highly experienced critics, are usually the result of insufficient knowledge of literature. Consequently knowledge always has the power of veto over taste.
Let us take the criticism of the new literature that appeared around 1922—recent enough for judicial criticism to be still relevant, but far back enough for us to see something of the academic process too. The judicial critics of this literature varied greatly: some of them denounced much of the best of it as morbid, precious, decadent, literary Bolshevism, and what not. They were wrong, obviously, but, if value judgments cannot be demonstrated, how do we know they were wrong? On the writers of this period whom we now take most seriously, such as Eliot and Joyce, a good deal of academic criticism has been done. The general direction of this academic criticism has been to show that writers who seemed in 1922 to be bringing something utterly new and unprecedented into literature were in fact deeply traditional writers. The rejecting critics went wrong, not through the failure of their taste or judgment, but through not knowing enough about literature. They did not understand the traditions of literature that made Ulysses and The Waste Land possible; they did not in consequence realize that literature could do this kind of thing too. It is out of the expansion of knowledge brought by the academic critics that the importance of Eliot and Joyce has been established, not through any contest of pro and con value judgments.
We have now to try to translate these general principles into axioms of procedure. In the scholarly techniques dealt with in previous essays there is a ratio between accuracy and limitation of perspective. In editing or bibliography there is a goal: one may do something in these fields that may be definitive, not needing doing again in the foreseeable future.Here, as Brahms said of a contemporary composer, one’s work may be immortal for quite some time. But in historical or literary criticism there is no possibility of being definitive: one can only do the best one can and cut one’s inevitable losses. At the same time the opposite extreme, expressed by such phrases as de gustibus non est disputandum and similar indications of too great an ascendancy of judicial criticism, is equally fallacious. The academic critic is, or should be, contributing positively to a growing body of knowledge; and if there is no goal for him, there is for the total work of criticism of which he is part.
The more a work of scholarship depends on the kind of statement of fact that can be definitely verified or contradicted, the narrower in range it is likely to be. In historical criticism accuracy of fact is subordinate to interpretation and organization of argument, but it is still sufficiently important to keep one limited to one or at most two special fields. These fields steadily broaden with growing seniority, but still they are always readily identifiable. The scholar’s work is highly specialized, and the expenditure of time and energy needed to keep up with everything published in any scholarly field is in itself prodigious. At the same time the scholar is likely to have many other literary interests: he may teach courses, or at least texts, outside his field; he needs more than a superficial knowledge of the major classics of literature that do lie outside it; he has presumably some interest in contemporary literature or in “what’s going on”; he may be a poet or novelist himself. In any event he will be a cultivated person with many more literary interests than his primary scholarship covers.
These peripheral interests become relevant as soon as he moves from scholarship to criticism. The first principle of academic criticism is that in it the critic is attempting to interpret something in the light of all the literature he knows. This will in fact include everything else he knows, and, if he has seen a great deal of life and profited by his seeing, even nonliterary experience may be relevant. But for the most part one’s experience of life has a minor role to play in criticism: it is one’s literary experience, and the training of the imagination and sensitivity that reading and writing alone can give, that is its basis. For criticism, wisdom and virtue are doubtful aids, and only the madness of much learning can be consistently trusted.
The primary understanding of any work of literature has to be based on an assumption of its unity. However mistaken such an assumption may eventually prove to be, nothing can be done unless we start with it as a heuristic principle. Further, every effort should be directed toward understanding the whole of what we read, as though it were all on the same level of achievement. We often use such phrases as “see what the author was trying to do,” and the like; these of course are inaccurate, but they do express something of this conception of an ideal or perfect product. We should hold to this conception as long as possible, in defiance of everything our taste tells us, even if the work we are studying is as obviously uneven in achievement as The Revolt of Islam or Oliver Twist. The critic may meet something that puzzles him, like, say, Mercutio’s speech on Queen Mab in Romeo and Juliet [1.4.55–94], and feel that it does not fit. This means either that Shakespeare was a slapdash dramatist or that the critic’s conception of the play is inadequate. The odds in favour of the latter conclusion are overwhelming: consequently he would do well to try to arrive at some understanding of the relevance of the puzzling episode. Even if the best he can do for the time being is a far-fetched or obviously rationalized explanation, that is still his sanest and soundest procedure.
Or, again, if a critic is working on Paradise Lost or The Prelude, his taste and discrimination as a reader are likely to tell him that parts of these poems are duller than others. But his taste should not be allowed to interfere with the serious work of criticism. For that, the fact that Milton or Wordsworth found these passages relevant to their total scheme is primary; the dullness of the passage itself is incidental. If the attempt to understand the poem leads to some explanation of why the passage is dull, then the estimate of value may come into its own. When as a student I first read the speech in book 3 of Paradise Lost in which “God the Father turns a school divine,” I thought it was grotesquely bad.10 I have been teaching and studying Paradise Lost for many years, and my visceral reaction to that speech is still exactly the same. But I see much more clearly than I did at first why Milton wanted such a speech at such a point. I also think I know one reason why I find the speech so unsatisfying: it is inconsistent with the Christian view that no one can know the Father except through the Son, and this makes it lack conviction, not because I or any other reader holds that view, but because Milton did. I am not giving my own reactions as models; I am merely trying to illustrate what seems to me the normal academic process. Finally, what holds for major literature holds also for minor literature. Ordinary fairness demands that every poet must be treated as though he were potentially first-rate, as far as the initial act of understanding is concerned.
What is true of the individual work applies to a writer’s entire work. Every poet writes good and bad poems, but again the critic’s primary task is to understand and explain, and a poet’s worst poem may be as relevant to the full understanding of him as his best one. Sometimes full understanding corrects stock value judgments that may have been parroted from the poet’s own day. Wordsworth is a good example: many a teacher proceeds as though the reactions of Wordsworth’s less perceptive contemporaries to The Idiot Boy or the original version of The Thorn were definitive. He may have been accustomed to ridicule these poems to his students, and to have his judgment confirmed by appreciative chuckles. If, however, he tried to understand, not only why Wordsworth wrote these poems, but why Wordsworth wrote them in the particular way he did, he might make a fresh critical discovery of his own, at a possible cost of injuring his popularity as a lecturer.
In what follows I take “poem” as representative of everything else in literature, because it is a short word. The process of academic criticism begins, then, with reading a poem through to the end, suspending value judgments while doing so. Once the end is reached, we can see the whole design of the work as a unity. It is now a simultaneous pattern radiating out from a centre, not a narrative moving in time. The structure is what we call the theme, and the identifying of the theme is the next step. By “identifying” a theme I do not mean spotting it: the theme is not something in the poem, much less a moral precept suggested by it, but the structural principle of the poem. In Robert Frost’s Mending Wall, for instance, the theme is not the question whether good fences do or do not make good neighbours: the theme is the identity of the “something” with which the poem begins. We are not told definitely what it is, except that it is not elves; but whatever it is, the contrast of the two human attitudes toward the wall and the two directions of the seasons, toward winter and toward spring, radiate from it as the centre of the poem.
The theme cannot be identified, in this sense, without revealing to us the general structure of the poem. This soon brings us to the ordinary structural units of literature, the images, which in works of fiction include characters. There are two aspects of imagery of particular importance. One is repetition, a study of which normally gives us the clue to the theme, as the repetition of images of water and fire leads us to the themes of The Dry Salvages and Little Gidding in the Eliot Quartets. The other is modulation, or the reappearance of an image in a different context. An example is the image of Nature hiding her guilty front with innocent snow in Milton’s Nativity Ode, reappearing later in the poem in the form “And leprous sin will melt from earthly mould” [l. 110].
After the study of the imagery, we move up still more closely to the “texture” of the poem, and give our attention to the ambiguities of the syntax and the complex meanings of the words used. Poetry is not a structure of words aligned with an external meaning, but a structure of words that contains meaning, hence such ambiguities and complexities may be functional to poetry, and so relevant to criticism. Analysis of this sort can become very intricate, hence I suggest a deductive process, starting with the central theme and moving from structure to texture, in order to keep the perspective of the poem in view, the unity to which everything else must be relevant. If, for example, I am studying The Ancient Mariner, it may occur to me that the Ancient Mariner is suffering from remorse. This in turn may remind me that remorse is connected etymologically with biting, hence “I bit my arm, I sucked the blood” [l. 160], and that it is connected in sound with remora, the fish supposed to be responsible for the becalming of ships. Here I am no longer in touch with The Ancient Mariner, but have wandered off into an associative fantasy of my own. Such irrelevance is not a hazard to an experienced critic, but it may be one to the beginner.
The critical techniques so far dealt with may be summed up under the heading of commentary. Commentary means, in general, the translating of as much as possible of a poem’s meaning into discursive meaning. Commentary is always allegorical, in the sense that it aligns the poem’s meaning with another structure of meaning to which it is related point by point. Commentary is also, when it deals with major literature at least, inexhaustible. Just as clergymen can preach sermons on the Prodigal Son to the end of time, so an infinite number of critical commentaries could be written on the plays of Shakespeare or the novels of Balzac. Such criticism is the lineal descendant of the allegorical commentaries on the Bible and Classical mythology which bulk so large in medieval and Renaissance culture. The critic should be aware that while commentary is an essential part of criticism, it is also relatively easy to write. If he is interested in a major writer, he will get critical ideas suggesting commentary from that writer: or at least if he cannot get such ideas from a major writer he should certainly be in some other line of business. It is also not difficult to set up analogical patterns between a writer’s work and the generally accepted ideas of his time, to see parallels between Shakespearean comedy and the mythical structure of Christianity, between Tennyson or Hardy and the emotional impact of Darwinism. But the critic has still much to do before he is in the central area of literary criticism.
What commentary does not do, by itself, is to identify the poem, which we said earlier [124] was the aim of academic criticism. To identify something means seeing it as an individual of a class, and criticism has not seriously begun until it puts the poem into its proper literary context, as one of a family of poems. We saw that criticism was neither purely historical nor purely contemporary in attitude, but a combination of both. We are now ready to take another step and see that it is a third category which unites these contrasting and often opposed attitudes. This category is literature itself, considered not as the bibliography of literature, not as the aggregate of plays and poems and novels that have got themselves written, but as an order of words, the study of which can be as systematic and progressive as the study of anything else. The sense that every work of literature has a context in literature makes criticism finite, not in the sense of limiting the critic, but in the sense of giving his work a direction and an aim. Nobody can write a definitive book on a major poet, but, we said, the sense that important criticism contributes to a consolidating structure of knowledge is very strong. Commentary alone cannot give any indication of what this larger structure is to which the critic is contributing.
First, every poem belongs to the total body of writing produced by its author. The more important the poet, the more obviously everything he produces will assume the appearance of a single larger poem. Reading in context is an essential principle of all reading, whether of literature or not, and to study, for example, Yeats’s Vacillation as a poem by itself, without comparing the similar contrasts of cycle and ascent in Dialogue of Self and Soul or Among School Children, or without following up similar ideas in A Vision and Rosa Alchemica, is simply to read the poem out of context. The next step takes us into an aspect of historical criticism. We cannot think long and deeply about, say, Keats’s Ode on Melancholy without feeling that its essential relationships cannot be confined to Keats, but that there is a larger reference which is one of the things that the term “Romantic” expresses. This larger reference makes the work of completely different poets, such as Shelley and Byron, or even French and German Romantics that Keats had never heard of, part of the context for Keats’s poem.
Criticism, in the limited sense used here, is, we said at the beginning, the attempt of a student of literature to interpret works of literature in the light of all the literature he knows. We are now, perhaps, in a better position to see how this is true. Suppose a student of literature is interested in Shakespeare. He will start by attaining some familiarity with the aspects of Shakespearean scholarship outlined elsewhere in this handbook: the struggles of editors with Quartos and Folios and Bad Quartos, the research on the structure and acting conditions of Elizabethan theatres, the evidence amassed for the order and dating of the plays. His bent for criticism, however, will soon take him toward reading Shakespeare against a background of other Elizabethan writers. The more he knows about sonnet conventions in Sidney and Spenser and their predecessors back to Petrarch and Dante, the less time he will want to waste in trying to ascertain what or who Mr. W.H. or the dark lady may have been outside literature. He will concentrate on the sonnets as poems, therefore as subject to the conditions of literary expression. In studying the tragedies, he will eventually come to find Shakespeare’s contemporaries less immediately relevant to his investigation than Sophocles or Aeschylus, not because Shakespeare drew from them, but because tragedy is a certain kind of literary structure, and the greater the writer of tragedy, the more clearly he exemplifies the structure and meaning of tragedy. Thus criticism, as soon as it disentangles itself from other forms of literary scholarship and evolves beyond commentary, finds itself preoccupied with two aspects of literature in particular: convention and genre.
With the next step we enter what for some unfathomable reason is a less explored area of criticism. Works of literature are not created out of nothing: they are created out of literature itself, so far as the poet knows it. Literature bears testimony to its own literary origin by being highly allusive, and allusions in literature are not stuck in: they are a part of the larger structure to which the poem belongs. We should be reading Lycidas out of context if we detached it from the rest of Milton’s work, but we should equally be reading it out of context if we detached it from Theocritus and Virgil, to whom the poem itself explicitly and repeatedly alludes. The structure of every poem is in one respect unique, but there is also a conventional element in the structure derived from the conventions and genres the poet is using. Lycidas is a pastoral elegy, and this structure is in it much as the conventional sonata form is in a Mozart symphony. We also find a conventional element in the imagery, particularly noticeable in such things as the “sanguine flower inscrib’d with woe” [l. 106], which turns up in every poem of this type from ancient Greece to our own day.
As we continue to study works of literature in the context of literature itself, as indicated by convention, genre, and allusion, something of the shape of literature as a total order of words begins to dawn on us. Literature associates, by words, the nonhuman world of physical nature with the human world, and the units of this association are analogy and identity, which appear in the two commonest figures of speech, the simile and the metaphor. The clearest forms of such association are in mythical images, where, for example, we have a “god,” who is human in shape and character and yet identified with something in nature like the sun or the sea. Looked at in this way, literature as a whole appears as the direct descendant of mythology, filling in many areas, especially in realism and irony, that mythology barely touches, but keeping the mythological sense of a panoramic view of the human situation, a perspective to which the greatest works of literature invariably return. This will enable us to understand why our inherited mythologies, Biblical and Classical, form by far the largest body of allusions in literature, and seem to be sitting in the exact centre of our whole imaginative heritage. Many critics, including the present writer, think of the conventional structures mentioned above as myths, because all of them derive from myths. Similarly, we may call the conventional images that recur throughout the whole of our literary experience “archetypes,” a word which has been used since Plato in the sense of a pattern or model used in creation.
We said that there is always some tension or polarity between what a work of literature meant to its own time and what it means to us now. One perspective does something to correct the other, but the two remain balancing one another more or less illogically and paradoxically. Eventually the historical perspective expands into the sense of tradition, where the writer’s own age is seen as part of a continuum, handing on what has reached it from earlier ages and reaching down to our own time. The contemporary perspective, similarly, expands from a judicial attitude into a scholarly or academic one, and by doing so reaches the counterpart of tradition, the sense of the total form or telos of criticism as the theory of literature, studying literature as a coherent and unified order of words, and becoming in itself a coherent and unified body of knowledge by virtue of that order. The historical and contemporary attitudes can never be really reconciled, but tradition and telos can be. This is the point at which criticism moves into the conception described by Matthew Arnold as culture, where the study of the best that has been thought and said becomes an organized force in society, dissolving its grosser inequalities, refining manners, disciplining the emotions as well as the intellect, and assimilating the actualities to the ideals of human civilization.11
One would naturally think that the social and moral benefits of studying literature come from the content of literature, from its power to express great thoughts with the appropriate emotional resonance. Yet we have been saying all through this paper that the primary axiom of critical procedure is: Go for the structure, not for the content. This must be the invariable attitude of every genuine critic, whether he is teaching Paradise Lost in a church college or on a witness stand testifying to the merits of Lady Chatterley’s Lover. We begin now to see how it is only the study of the structure of literature that can lead to culture in Arnold’s sense. For we soon discover that structure is not self-contained, that individual works of literature are not locked up in windowless monads separated from each other, but that there are family likenesses resembling the species, genera, and phyla of biology. Eventually we come to the point at which the form of literature as a whole becomes the content of criticism as a whole. It is here that we begin to be interested in larger questions: why man produces literature, what it does for society, what its connections are with other uses of the mother tongue.
It is perhaps worth remarking, in conclusion, that, while literature and scholarship have been around for quite a time, criticism in the sense used here has hardly begun. The development of criticism is bound to lead—has already led—to an enormous increase of self-consciousness in literature itself, and not all of that will be good. But for better or worse, our civilization, if it survives at all, will be one in which criticism and literature, that is, the theory and practice of literature itself, will be two parts of one thing.