Psychoanalysts differ widely among themselves over which aspects of Freud’s theories they wish to remember and commemorate. Freud’s theories, so far from constituting a unitary, fixed structure, which either stands or falls as a whole and which analysts subscribe to in its entirety, are really more a collection of miscellaneous ideas, insights, and intuitions, which Freud propounded over a span of fifty years, which he derived from three disparate sources—his clinical experience, his self-analysis, and the biological theories current in his lifetime—and which have proved capable of development and elaboration in several different and apparently, perhaps even actually, incompatible ways. That this last statement is true is shown by the fact that there exist today in Great Britain at least three different schools of psychoanalysis, all claiming to be Freudian and all capable of showing that their ideas can indeed be discovered, albeit often in embryonic form, somewhere in Freud’s writings.

Although the fact that Freud’s writings have proved to be more a quarry than an edifice is not my subject here, it does make it necessary for me to state explicitly which aspect of Freud’s thinking I consider most important and which particular group of Freudian concepts I shall be using while developing my argument.

It so happens that my own view in this matter coincides with Freud’s. To his and my mind the most important, the most seminal, and the most revolutionary idea that Freud ever had was his idea that the human mind is capable of thinking in two different ways or modes; that there are, to use the title of one of his papers, “two principles of mental functioning,” one of which is characteristic of our waking life, the other characteristic of dreaming and neurotic symptom-formation; and that it is possible to define and describe these two modes by presenting them in antithesis to one another, so that each can be conceived of as possessing characteristics which are the opposite of the other’s.

To my mind misleadingly, Freud termed these two antithetical principles or modes the primary and secondary processes, the primary processes being those characteristic of our dream life, the secondary processes being those characteristic of our waking thought. The primary processes, in Freud’s terminology, are condensation, by which mental images tend to fuse with one another, and displacement, by which they tend to replace and symbolize one another; these processes ignore opposites and the categories of space and time, and are wish-fulfilling.

The secondary processes, on the other hand, respect the differences among images, obey the laws of grammar and formal logic, take cognizance of opposites and of the categories of space and time, and are adapted to the “realities” of the external world. To use a terminology which Freud himself did not use, the primary processes are iconic and nondiscursive, the secondary processes are verbal and discursive. The meaning of primary process utterances—if one can call a dream or a symptom an utterance—is ascertainable only by teasing out items and threads from the total agglutination, amalgam, or fuzz that constitutes the dream, the meaning of secondary process utterances is ascertainable by reference to the dictionary definitions and syntactical rules of the language which the waking speaker or writer is using. Furthermore, it was Freud’s view that the primary processes in some sense come before the secondary processes—hence, of course, the nomenclature—and are, therefore, more infantile, more primitive, and less adaptive than the secondary processes, which are learned during the course of each person’s development from being a primitive, fantasizing infant to a civilized, adapted adult.

Now, as a method of approaching the main theme of this essay, I intend to make a number of introductory comments on Freud’s idea, or discovery, of the existence of these two principles of mental functioning.

First, Freud unwittingly got himself involved in an inherently paradoxical activity when he tried to formulate in words the nature of a type of thinking which is essentially nonverbal and which is, therefore, of necessity falsified by being put into words. But as a rationalist and a scientist, he really had no option but to try, since making use of already available techniques for handling ineffable experiences and phenomena which cannot be verbalized, such as the Via Negativa of the mystics and the apophatic mode of argument of the theologians, would have seemed to him to be a betrayal of his ideal, the creation of a truly scientific psychology, and of one of the basic assumptions of his generation, the assumption that the only real truths are scientific truths.

As my last paragraph will have been obscure to those who are not well versed in theology, I must, I think, make its point a second time and in another way. Freud asserted that one of the major characteristics of the primary processes is that they are unconscious, and in his initial formulations he located them in a fictive space or area, which he designated negatively as the “unconscious.” And even when he changed this area’s name to the id, his description of it consisted largely of what it is not.


It is the dark inaccessible part of our personality; what little we know of it we have learnt from our study of the dream-work and of the construction of neurotic symptoms, and most of that is of a negative character and can be described only as a contrast to the ego. We approach the id with analogies; we call it a chaos, a cauldron…. It is filled with energy reaching it from the instincts, but it has no organization, produces no collective will…the logical laws of thought do not apply…contrary impulses exist side by side without cancelling each other out…. There is nothing in the id that corresponds to the idea of time…. The id, of course, knows no judgements of value; no good and evil, no morality. [Freud, New Introductory Lectures on Psychoanalysis, No. 31]

Now, so far as I know, the only other example of anyone proclaiming the existence of something profoundly important by asserting that it can only be approached with analogies, and that it is not anything that can be defined and asserted positively, is to be found in negative or apophatic theology, which argues that human language, when applied to God, is inevitably inexact and that it is therefore less misleading to use negative language about God than positive—“to refuse to say what God is, and to state simply what He is not.” (See Timothy Ware, The Orthodox Church, Penguin, 1963.) Curiously enough, the one positive attribute of God which negative theologians assert can be known about God is His energy, which again corresponds to Freud’s account of the id. However, my reason for mentioning negative theology is not to discuss the parallels that can be drawn between the theological concept of God and the psychoanalytical concept of the id, interesting though it might be to do so, but to prepare the ground for arguing later that there is something about the literary imagination that can only be stated in negative terms.

Secondly, Freud formulated his theory of the primary and secondary processes in a cultural milieu very different from our own. Freud was born in 1856, he was fifty in 1906, and his formative years were, therefore, pre-Einstein, pre-Picasso, pre-cubism, pre-Ezra Pound, pre-James Joyce. As a result it was natural for him to assume a much closer relationship between the verbal, the rational, and the realistic on the one hand and the nonverbal, the irrational, and the imaginary or imaginative on the other hand than any thinker can today.

Freud formulated his ideas before the visual arts had ceased to be representational and before literature had begun to explore the possibilities of fragmenting and manipulating the syntactical structure of language. He was therefore able to assume that when painters painted pictures they were depicting objects and scenes that could, in principle, also be described in words, and that when writers wrote books they were using language in the same sort of way as scientists do when they write learned treatises. As a result there appeared to him to be—and indeed given the historical context there was—nothing incongruous in assuming that of the two types of mental functioning, one, the verbal, rational mode, was characteristic of the ego, of consciousness, of health, of successful adaptation to reality, and the other, the nonverbal, irrational mode, was characteristic of dreamers, neurotics, lunatics, infants, and primitive people; and that the capacity to use the former was dependent on repression of the latter.

The idea that the primary processes are unconscious, primitive, neurotic, archaic, etc., and are normally subject to repression was to cause psychoanalysis considerable trouble, both in its theorizing and in its public relations, since it soon became evident that there was some similarity between the imaginative activity displayed by artists and writers, and the primary processes described by Freud as characteristic of dreaming and symptom-formation. Given the clinical origins and bias of psychoanalysis, the easiest, the most tempting way of explaining this similarity was to assert that artists and writers are neurotic and that works of art are analogous, or homologous, to dreams and neurotic symptoms; and that the techniques of psychoanalytical interpretation can be transferred without modification to artists and their works.

This idea is still around today, despite Freud’s rather belated disavowal of it in 1928, when he wrote, “Before the problem of the creative artist analysis must, alas, lay down its arms.” It owes its vitality, I suspect, to four disparate sources. First, to envy—we are all a bit envious of creative people and it is comforting to entertain the idea that they aren’t after all so very different from ourselves and that their gifts may not truly be gifts but by-products of their neuroses. Second, to the pecking order that exists among academics and the intelligentsia generally—if scientists, and psychoanalysts almost universally claim to be scientists, could explain creative activity, could analyze the literary imagination, and could slot it neatly into the scientific scheme of things, they would have one-upped the artists and increased their own prestige.


Third, to the rise of English Lit. as an academic study and the resulting expansion of the PhD industry, which has led to a demand in academic circles for new techniques to apply to the victims of literary thesis-writers—and what is more tempting than to add to literature’s own critical armory a few, mostly character-assassinating, weapons borrowed from psycho-analysis. And fourth, the fact that there is indeed some similarity between the creative imagination and the oneiric, dream-producing faculty, a similarity long recognized by artists and writers themselves, and that it is, therefore, legitimate to discuss what the nature of this similarity is.

But, and it is a big but, discussion of this similarity is queered from the onset if one assumes, as most of the early analysts seem to have done, that in healthy persons the primary processes are repressed and that people in whom they are not repressed are ipso facto neurotic. The whole idea that artists and writers are neurotic stems, I am suggesting, from one simple but fallacious assumption, that the primary and secondary processes are mutually antagonistic, and that the former have, in health, to be relegated by repression to a curious underworld, the id or the unconscious. If this were true, then painters who can imagine what they intend to paint with quasi-hallucinatory, eidetic vividness and writers who can conjure up characters who seem to take on an independent life of their own would indeed be neurotic and psychotic.

But if one starts from another assumption, that the primary and secondary processes co-exist from the beginning of life and that under favorable conditions they may continue to function in harmony with each other, one providing the imaginative, the other the rational basis of living, then creative people may be conceived to be those who retain into adult life something of that imaginative freedom which healthy children display openly but all too many grownups in our present rationalist, bourgeois culture lose when they enter the adult world. According to Freud, “A sharp and final division between the content of the two systems does not, as a rule, take place till puberty” (Freud, The Unconscious, 1915, Standard Edition vol. 14., p. 195).

It may seem that I am being perverse in first asserting that in my opinion Freud’s most important contribution was his formulation of the distinction between the primary and secondary processes and then going on to qualify this statement by arguing that in one vital respect his formulation was misguided and requires modification. I must, therefore, defend myself briefly from this charge by making three points.

First, hints of the possibility of modifying Freud’s theory in the direction I am suggesting can be found in Freud’s own writings, particularly if one allows oneself to reverse statements he made, e.g., his statement that “a total severance of the two systems is what above all characterizes a condition of illness” (Freud, The Unconscious, 1915, SE vol. 14, p. 194). Second, I am far from being alone among analysts in rejecting the idea that the primary processes are archaic and maladaptive; in Great Britain D.W. Winnicott, John Bowlby, Marjorie Brierley, and Marion Milner have all in their different ways done so.

Third, attempts to reconcile Freud’s position with a positive, nonreductionist attitude toward imaginative activity always seem to lead to formulations which are tortuous and obscure but still leave the central mystery unexplained. For instance, Heinz Hartmann, the distinguished American psychoanalyst, who is on record as having subscribed to “the theoretical ideal of rational action” and to the view that healthy development leads to “alienation of the id from reality,” was forced into using such phrases as “regressive adaptation,” “detours through the archaic,” and “regressions in the service of the ego” when writing appreciatively about art. The attempt to explain imaginative activity by a theory primarily designed to explain neuroses without impugning either its value or its healthiness seems to me to amount to no more than asserting that creative people possess some knack for getting round the repressions assumed by the theory to be normal and then finding some impressive-sounding verbal formula to describe this knack.

Freud’s conception of the existence of unconscious primary processes was, according to Freud’s own view of the nature of science, profoundly unscientific. The unscientific nature of this part of his theory seems only to have struck him forcibly when he came to consider the actual details of the imagery used by the primary processes and came to recognize the importance of symbolism in general and of sexual symbolism in particular. It is not, I think, generally known that the original 1900 version of The Interpretation of Dreams had no section on symbolism in it and cites only one dream exemplifying sexual symbolism, that it was only in the 1914 edition that a section on symbolism appeared for the first time, and that in the present standard edition that section accounts for only fifty-five of its 623 pages.

Of course the embarrassing, unscientific nature of symbolism arises from the fact that symbolic interpretations can be arrived at intuitively and, as Freud himself put it, intuition “is exempt from all criticism and consequently its findings have no claim to credibility” (Freud, The Interpretation of Dreams, SE vol. 5, p. 350). Although this is a very respectable scientific opinion, it is hardly applicable to the subject of symbolism, since symbolism is a form of metaphor and, as Aristotle said, “a good metaphor implies an intuitive perception of the similarity in dissimilars” (Poetics, chapter 22).

As a result of Freud’s understandable rejection of intuition, objective evidence in support of symbolic interpretations of dream imagery had to be culled from sources remote from the natural sciences on which Freud had hoped to build psychoanalysis; he had to turn to myths, to folklore, to anthropology, to etymology, to dirty jokes, all “soft” subjects from a natural-scientific point of view.

If Freud had lived today, he would not, I think, have had to be embarrassed by the apparently nonscientific nature of this aspect of his discoveries, since the emergence of linguistics as a scientific discipline would have enabled him to use that science as a model instead of neuro-anatomy. Instead of constructing, as he did, a mental anatomy in which the primary processes were located in one part of a mental apparatus and the secondary processes were located in another, he could have formulated a para-linguistic science, which might perhaps have been called oneirics, with iconic, structural, and semantic branches. He might have formulated sets of rules governing both the translation of oneiric, iconic statements into phonetic, verbal utterances and the setting up of obstacles against translation.

In other words, the rules would have to be able to explain why, and under what conditions, dreams can be understood and interpreted to the mutual enlightenment of both interpreter and analysand, and why, and under what conditions, they sometimes cannot be. Another set of rules would have had to account for the fact that imagery related to biological destiny, which includes birth and death as well as sex, seems to occupy a central position in oneiric structure—and perhaps in language structure too. (See my article “Is Freudian Symbolism a Myth?” in The New York Review, January 24, 1974, and in Symbols and Sentiments, Academic Press, 1975.)

A few paragraphs back I suggested that there are indeed similarities between the literary imagination and the faculty of dreaming, and the time has come, I think, for me to discuss in detail what these similarities are. I must, however, first emphasize that there never seems to have been any doubt among poets and writers themselves that some important connection exists between their creative activity and dreaming. It would be possible to |cite innumerable examples of writers using their dreams as the initial source of their inspiration, of their claiming, sometimes untruthfully, to have composed poems in their sleep, of their including dreams they have had in the text of their novels, of their claiming specifically to be “dreamers of dreams.” These facts are, however, so well known that it would be mere padding on my part if I were to spend time citing them, so instead I shall give just one quotation from Charles Darwin.

The Imagination is one of the highest prerogatives of man. By this faculty he unites former images and ideas, independently of the will, and thus creates brilliant and novel results. A poet, as Jean Paul Richter remarks, “who must reflect whether he shall make a character say yes or no—to the devil with him; he is only a stupid corpse.” Dreaming gives us the best notion of this power; as Jean Paul again says, “The dream is an involuntary act of poetry.”

This quotation is from Darwin’s The Descent of Man, which was first published in 1871. Darwin borrowed the quotations from Jean Paul Richter from a book by Henry Maudsley, the English psychiatrist who provided the money with which the London County Council built the Maudsley Hospital. Note that both Darwin and Richter emphasize imagination’s independence of the will, a point to which I shall return later.

The first and most striking similarity between dreams and works of literary imagination is that they can be granted or refused meaning according to the predilection of their viewer, hearer, or reader—English seems to lack a word which would embrace a person viewing his own dream, a person listening to an account of someone else’s dream, and a person reading a poem or novel. Now, this statement is obviously true in the case of dreams. Despite Freud, it is probably still true that the majority of educated people in the United Kingdom, though perhaps not in the United States, do not attribute any meaning to dreams, and, on the face of it, it is a matter of personal bias or temperament whether anyone sees meaning in dreams or regards them as, say, analogous to the “noise” made by electronic equipment when it is switched on but not actually working.

It is, however, less obvious that it is possible to deny meaning to poems and novels. I expect we can all think of acquaintances who are literate but literal and not literary, who don’t really get the point of novels and poems, and are flummoxed by the fact that the realm or order of reality to which they should be assigned is not readily defined or located; who are flummoxed by precisely that element of fiction in fiction which makes it impossible to read novels simply as disguised biography or disguised sociology, and who don’t really understand why poems need to be laid out on the page in such a wasteful, extravagant manner.

There would, I think, have been something facetious about my having drawn attention to the fact that dreams, poems, and novels can all be regarded as meaningless, were it not for the fact that if one does attribute meaning to them they immediately acquire not a single meaning but multiple and manifold meanings. Unlike factual statements, like “The Battle of Waterloo took place on June 18, 1815” or “Arsenic is a poison,” which have only one meaning, poems, novels, and dreams either have no meaning or several meanings. Once one has recognized or decided that they have meaning, they become open to interpretation, and characteristically several not mutually exclusive interpretations can be made of them. And characteristically too, exegesis of these interpretations takes up more space, and uses more words, than does the dream, poem, or even the novel itself.

This is due, of course, to the fact that the imagination unites, to use Darwin’s word, or condenses, to use Freud’s word, numbers of images and themes into a unitary whole, which therefore takes up less space than the individual items do if enumerated in series. I could mention two examples. The first dream that Freud ever “submitted to a detailed interpretation,” that of Irma’s injection, which he himself dreamed in 1895, took him two-thirds of a page to recount and fourteen pages to interpret. And second, in a volume of essays entitled Interpretations, edited by John Wain, it takes Dennis Ward thirteen pages to answer the question “What did the sonnet ‘The Windhover’ mean to Gerard Manley Hopkins?,” i.e., just under a page to a line. Furthermore, his answer involves him in mentioning at least ten verbally distinguishable meanings for one of the poem’s two central images, the falcon—the other central image is air or wind—even though these ten meanings unite in the poem itself to form one meaning which it is artificial to dissect into sub-meanings.

Interestingly enough, neither Freud nor Ward suggests interpretations of a kind that we nowadays think of as being specifically psychoanalytical, and Freud’s interpretation of his own dream does not include any mention of any underlying over-all meaning, but anyone reading it today with hindsight can see that it has one, though it is not clear, to me at least, whether or not Freud saw this meaning and suppressed it for reasons of tact and delicacy. I shall have reason to return to Hopkins’s “The Windhover” later.

Although I understand it is accepted practice for literary critics to proffer multiple but interconnected meanings for literary works they interpret, psychoanalysts are commonly criticized for doing precisely the same thing when they interpret dreams—and are indeed sometimes themselves unhappy about doing so. The difficulty here is, I think, based on a confusion between causes and meaning. If one asserts that the cause of A is X, then one is in difficulties, though not always insuperable ones, if one also asserts that its cause is Y and Z, but if one asserts that a meaning of A is X, there are no objections to asserting that it also means Y and Z, since there is no rule that utterances, other than scientific statements, should have only one meaning.

Unscientific utterances can, and indeed usually do, have double meanings, implied meanings, unintended meanings, and can hint and insinuate, and may indeed mean the opposite of what they apparently mean, especially if they are said in a certain tone of voice. Here again we have an example of psychoanalysis being caught in a trap created by its own history, of laying itself open to attack for being unscientific, when it has in fact entered a field where it has no need to claim to be scientific.

A third similarity between dreams and imaginative works is that their production is independent of the will. We do not make up or construct our dreams; they occur or happen to us, and while we are dreaming we do not recognize the dream as a creation of ourselves. The same seems to be true of at least the initial idea or impetus of any literary work; it just comes to the writer or poet. A necessary precondition of all imaginative activity seems to be what Keats called “negative capability,” the ability to allow oneself to be “in uncertainties, mysteries, doubts, without any irritable reaching after fact and reason”—in other words, the ability to abandon Hartmann’s “theoretical ideal of rational action” and to stop trying to “master reality” in favor of letting oneself happen. And, at least for some writers, the execution as well as the conception of a work of art may also be largely independent of the will. Coleridge claimed that “Kubla Khan” came to him in his sleep, though he must have decided after awakening that it was worth writing down, and Enid Blyton, author of the Noddy books, has left a vivid account of how her characters took over while she was writing her books.

“I shut my eyes for a few minutes, with my portable typewriter on my knee—I make my mind a blank and wait—and then, as clearly as I would see real children, my characters stand before me in my mind’s eye. I see them in detail—hair, eyes, feet, clothes, expression—and I always know their Christian names but never their surnames…. I don’t know what anyone is going to say or do. I don’t know what is going to happen. I am in the happy position of being able to write a story and read it for the first time, at one and the same moment…. Sometimes a character makes a joke, a really funny one, that makes me laugh as I type it on my paper—and I think, ‘Well, I couldn’t have thought of that myself in a hundred years!’ And then I think, ‘Well, who did think of it then?”‘ [Quoted from Barbara Stoney, Enid Blyton: A Biography, Hodder, London, 1974]

Although I have no particular wish to sing Enid Blyton’s praises, it must be admitted that her question “Well, who did think of it then?” is an extremely good one, and that she was far from being what Jean Paul Richter called a stupid corpse who had to reflect whether she should make a character say yes or no. This curious fact that the literary imagination is independent of the will, of the self-conscious ego, is presumably the reason why poets and writers before the rise of psychology could believe literally in their inspiring Muse. It is also the reason why we do not accord full artistic authenticity to works of art which strike us as contrived or voulu.

But despite these similarities between the ways in which the imagination manifests itself in dreams and in works of literary art, there are, of course, numerous differences. Quite apart from the obvious fact that most dreams are more like moving pictures than the printed word, dreams are in general less organized, less unified, indeed less condensed than works of art; they more often resemble someone who is groping for the appropriate metaphor than someone who has found it. If one accepts Richter’s dictum, quoted earlier, that dreams are involuntary acts of poetry, one has to add that they are, usually, also uncompleted acts of poetry. Something would still need to be done to them before they could be transferred from the private sector of experience to the public, before they could acquire universality.

This something-still-to-be-done must, I think, be something more than just the translation of visual, nondiscursive imagery into verbal, discursive language; in other words, the transformation of a dream or unconscious fantasy into a work of literary imagination cannot, I think, be analogous to the process of interpretation as practiced by psychoanalysts; it must rather comprise the casting of the central meaning in symbols which are part of the shared iconography of the culture of which the poet or writer is a member, and which, therefore, carry a heavy charge of shared, public associations and resonances.

Gerard Manley Hopkins’s “The Windhover,” for instance, fuses images derived not from his private experience but from ornithology, falconry, skating, chivalry, and Christianity—all public images with which any literate person could well be familiar—to make a personal statement about the relationship between divine inspiration, symbolized by the wind, and human aspiration, symbolized by the hovering falcon. Interestingly enough, if it were permissible to equate Hopkins’s God with Freud’s id, it would be possible to interpret “The Windhover” as a statement about the dynamic interdependence of the ego and the id, though a lot would be lost by doing so.

But I have stated my last point positively when I should have stated it negatively. It is not that the poet or writer actively masters the iconography of his times in order to be able to universalize his private emotions, but that one aspect of his “negative capability” is an exceptional sensitivity and receptivity to the iconographical network that constitutes the culture of his time, that makes it natural for him to express his private emotions in universal terms—or, indeed, perhaps not to distinguish between the individual and the universal, between the microcosm and the macrocosm.

I should perhaps here state explicitly something that will, I suspect, already have become clear: that I have no wish to dilate on the psychodynamics of the literary imagination. I must, however, state my impression, my conviction, that people who possess negative capability to a high degree seem not to conceive of themselves as opposed to an alien environment which they have to master by “irritable reaching after fact and reason,” but rather as a part of the universe which is capable of absorbing the whole into itself and then re-creating it by distillation in imaginative works; in other and psychoanalytical words, their relationship to “external reality” remains identificatory, without any drawing of impermeable ego boundaries between themselves and other people and other things. And secondly, they seem to be refreshingly free from the conventional notion that activity is masculine and passivity is feminine, and can therefore oscillate between active and passive states of being without feeling that their identity is threatened by doing so. As a result they can, for instance, imagine themselves into characters of the opposite—or rather other—sex as readily as into characters of their own.

This identificatory relationship with the outside world and this freedom from the conventional masculine-active and feminine-passive dichotomy can, of course, be used as evidence in favor of the view that artists are neurotic, and no doubt has been, but it could equally plausibly be used to prove the opposite—since drawing ego boundaries between oneself and others could well be regarded as a symptom of alienation, and lack of acceptance of the masculine-active feminine-passive dichotomy could well be regarded as a healthy immunity to indoctrination by cultural prejudices. It all depends on which part of the Freudian quarry one chooses to dig.

Although I have here, as indeed throughout this essay, emphasized the healthy core of imaginative activity, I have, of course, no wish to deny that mental creative processes may be interfered with by inhibitions, symptoms, and anxieties analogous to those which may impede biological creative activities. But it is, I think, as much an error to suppose that these disturbing factors are an essential part of imaginative, creative activity as to suppose that sexual hang-ups are an intrinsic part of orgastic capacity, a capacity which can indeed itself be conceived to be a form of negative capability.

Those who have followed my argument so far will appreciate that I think that there are intrinsic, inherent limits to the amount that can be said about the imagination in general and that much of what can be said can only be stated in negative terms; hence my initial reference to negative theology and my citation of Freud’s statement that most of what we know of the id is of a negative character and can only be approached with analogies. I should now like to end by giving three reasons for thinking that there are three inherent, intrinsic limits to the amount that psychoanalysts in particular can say about the specifically literary imagination.

First, literary studies seem to be largely concerned with questions of value and quality, i.e., with the question of why, and in what sense, some poems and novels are better than others, and psychoanalysis, to the extent that it remains attached to Freud’s ideal of a scientific psychology, can have nothing to say about value. Erik Erikson, the psychoanalyst who has most concerned himself with problems of meaning, value, and ethics—even to the extent of introducing the word “virtue” into his psychoanalytical vocabulary—has, as Daniel Yankelovich and William Barrett point out in Ego and Instinct (Vintage, 1971), been more engaged in “rediscovering prescientific truths” and absorbing them into psychoanalytical theory than in extending psychoanalytical theory so that it can contribute anything new to history, literature, and moral philosophy.

And, as Yankelovich and Barrett also point out, Erikson has consistently “blurred the extent of his divergence from the psychoanalytical movement” and the extent of his dependence on a philosophy at odds with that of Freud. Although I sympathize with Erikson’s motives and agree with much of what he says, it must, I think, be admitted that the creative traffic goes in the opposite direction to that in which it often appears to go, and that psychoanalysts have more to learn from historians, literary critics, and philosophers than they have to teach them.

Secondly, since imaginative activity is a classic example of the self as agent, to use John Macmurray’s phrase, accounts of the self that dreams, imagines, and creates are inevitably vitiated by the fact that the dreaming, imagining, creating self is not open to inspection or introspection. If the self tries to observe itself while creating, it inevitably fails, since the self-as-agent must willy-nilly become located in the observing, introspecting self and not in the part of itself that it is trying to observe; in other words self-scrutiny and negative capability are mutually exclusive states of mind.

Another way of putting this point is to say that the self that dreams, imagines, and creates is intrinsically nominative and can only be the subject of verbs, can only be “I” and never “me,” and that it does a disappearing trick if one tries to push it into the objective, accusative position. It is indeed somewhat similar to the concept of time, of which St. Augustine wrote that he knew what it was so long as no one asked him what it was, but that if he was asked, he didn’t know. (I owe this comparison to C.O. Evans, The Subject of Consciousness, Humanities Press, 1970.)

Third, it is, I think, open to doubt whether there is such a thing as a literary imagination which can be isolated and differentiated from other forms of imagination. My reason for this doubt stems from two sources; first, from the fact that our original knowledge of language is not of seeing words and sentences printed on paper but of hearing them spoken—for instance, our first encounter with poetry is hearing nursery rhymes recited or sung to us by our mothers. As a result there must, it seems to me, be a phonetic and musical substructure to the literary imagination, which must be an elaboration or special case of some wider form of imagination that is essentially auditory and musical, and I suspect that this is as true of prose as it is of poetry.

Moreover, since literature, like music, is a discursive art which proceeds through time, the literary imagination must, I suspect, contain some reflection of the rhythms and harmonies which the body uses to sustain itself as an organized entity through time. According to Walter Pater, “all art constantly aspires towards the condition of music,” and Henry Maudsley, whom I mentioned earlier, must have had a similar thought in mind when he stated that the roots of the imagination lay in “the multitudinous infra-conscious vibration of organic nature” (quoted by L.S. Hearnshaw in his A Short History of British Psychology, Methuen, 1964).

This Issue

April 3, 1975