The genius of the dream

  • Published on

  • View

  • Download



    Stanley R. Palombo

    At the climax of A Midsummer Night's Dream, Demetrius is released from his transferential infatuation with Hermia. This happens when he awakens from a dream that has successfully matched his current feelings for Hermia with a repressed libidinal fantasy of childhood. This example illustrates how condensation in dreams functions adaptively in matching a new expe- rience with previously stored representations of related events in the past. It also illustrates the ability of the matching process to go beyond the nar- row logical categories of waking thought to reach deeper levels of experi- ence otherwise inaccessible to the dreamer. This ability accounts for the important role played by dreaming in the creative process generalKy and in the day-to-day working-through process of psychoanalytic therapy.

    Freud's idea that dreams are an inner dramatization of the dreamer's unfulfilled wishes is, of course, much older than Freud. We find it in the work of the Greek tragedians (1) and in Shakespeare's plays. Near the end of A Midsummer Night's Dream (2), for example, Theseus offers what we recognize as the Freudian viewpoint on the tales told by the young lovers about their adventures of the previous night in the woods outside Athens:

    Such tricks hath strong imagination, That if it would but apprehend some joy, It comprehends some bringer of that joy. Or, in the night, imagining some fear, How easy is a bush supposed a bear! (V,1, 18-22)

    In our post-Freudian language, the tales are fantasies of wish fulfillment, psychodynamically indistinguishable from dreams. Like dreams, they reflect the drive states of the dreamers, rather than the actual events they have experienced. But Theseus's new bride, Hippolyta, feels that this inter- pretation does not do justice to the lovers' tales. She says:

    Stanley R. Palombo, M.D., Clinical Associate Professor of Psychiatry, George Washington University.

    The American Journal of Psychoanalysis Vo]. 43, No. 4, 1983 1983 Association for the Advancement of Psychoanalysis


  • 302 PALOMBO

    But all the story of the night told over, And all their minds transfigured so together, More witnesseth than fancy's images, And grows to something of great constancy; But howsoever, strange and admirable. (V,1, 23-27)

    There are many indications that Shakespeare shares Hippolyta's position in this controversy. Most importantly, her view follows naturally from the psychological movement of the play, which turns on the resolution of Demetrius's neurotic transference to Hermia through the therapeutic effect of a dream experience. The emotional climax of the play occurs when Demetrius becomes fully conscious of this change. The realization comes to him as he replies to Theseus's questioning, just after awakening from the night of enchantment:

    I wot not by what power- But by some power it is-my love to Hermia, Melted as the snow, seems to me now As the remembrance of an idle gaud Which in my childhood I did dote upon; And all the faith, the virtue of my heart, The object and the pleasure of mine eye, Is only Helena. To her, my lord, Was I betrothed ere I saw Hermia; But like a sickness did I loathe this food: But as in health, come to my natural taste, Now I do wish it, love it, long for it, And will for evermore be true to it. (IV, l, 163-175)

    Shakespeare only hints at the nature of the power that cured Demetrius. But he seems to be suggesting that, in order to understand it in depth, his readers would need a dream theory more resonant with Hippolyta's intui- tion than with Theseus's traditionally Freudian view.

    For those of Shakespeare's readers who have grown up in the Freudian era, this may seem a bit puzzling. But an examination of Freud's basic assumptions will help us understand why his theory of dream construction may not be adequate to meet Hippolyta's poetic challenge. Some of these assumptions have been cast into doubt by the laboratory research con- ducted in recent years on sleep and dreaming. Others have led to inconsis- tencies internal to the theory itself.

    Freud believed that a dream occurs when a repressed impulse succeeds in forcing itself through the repression barrier that bounds the Uncon- scious. This idea is very difficult to reconcile with the finding that dreaming


    occurs all night long in a repetitive pattern of 90-minute cycles. Whatever the drive state of the dreamer, his or her dreaming will occupy a period of about 20 minutes at the end of each one of these cycles. The pattern of our dreaming does, in fact, "grow to something of great constancy."

    The findings of the sleep laboratory are supported by the evidence pro- vided in Freud's analysis of his own dreams. When we examine the dreams and Freud's interpretations of them, we find that the meaning of each dream derives less from a single identifiable repressed impulse than from the network of associations in which it becomes embedded during the interpretive process. The larger this network becomes, for any given dream, the less plausible is Freud's hypothesis that the dream represents the fantasied fulfillment of an individual wish.

    In the body of Freud's dreams taken together (3), we find that the asso- ciative networks belonging to individual dreams ramify indefinitely as further associations emerge during his self-analysis, until ultimately they intersect to form a single extended network. It is difficult to imagine how portions of this seemingly endless web could be cleanly detached from it to form what Freud called the "latent contents" of individual dreams.

    More importantly, the existence of this extended associative network cannot easily be accounted for by Freud's theory of dream construction. According to that theory, isolated clumps of imagery are sought out by packets of repressed libidinal energy to be used as "vehicles" for their emergence into consciousness. The relationships among these complexes of imagery would have to be created de novo while the dream is taking place, as the repressed impulse is deflected by the censorship mechanism from one unacceptable complex to another.

    The problem here is that at each deflection the repressed impulse is called upon to make a series of totally implausible evaluations and decisions. It must search through some vast array of possible alternatives to the rejected complex and then select a substitute that qualifies as a suitable vehicle for the expression of the wish it embodies. At the same time, the substitute must differ sufficiently from the original wish to escape the scrutiny of the censor.

    But the "impulse" in question is by definition blind and primitive, intent only on finding a channel for the discharge of its quantum of energy. We are told that it is oblivious to the variety of obstacles to its fulfillment pre- sented by the external world. How then could it deal with the subtleties involved in selecting the proper disguise to fool the dream censor?

    The multiple connections linking together so many of Freud's dreams indicate that the associative network must exist prior to the construction of any individual dream; that the associative network must be, in fact, the structure of memory itself. It is a structure ordinarily inaccessiblle to our

  • 304 PAI.OMBO

    waking consciousness. But dreaming makes it partially accessible, and the individual dream contains the tracing of an exploration through a connected portion of it.

    Freud's theory could be modified to allow the deflected impulse to find its substitute imagery by following these preexisting pathways through the associative network. This modification would provide a reasonable picture of the mechanism of displacement as he described it. But it would not do as well with the more interesting process Freud called condensation.

    According to Freud, condensation occurs when two or more repressed impulses, each lacking a sufficient charge of energy for an assault on the repression barrier, succeed in combining their forces. This happens if their associated image complexes can be superimposed in such a way that similar nonthreatening elements of the imagery reinforce each other, while dis- similar threatening elements conflict and are canceled out. The net effect of such a combination would be that the energy quanta of the impulses are added together while their condensed imagery is rendered less threatening to the censorship mechanisms.

    Freud's theory would predict that the image complexes brought together in a condensation are selected through a series of more or less random colli- sions. What we actually see is that a condensation selectively superimposes newly recorded representations of current experience onto previously stored representations of experience from the often remote past (4).

    Representations of current experience are what Freud called "day resi- dues." He thought that every dream must contain such a residue, an "innocu- ous" fragment of daytime experience required somehow to convey the repressed impulse into consciousness. The supposed innocuousness of the day residue is not apparent in his own analyzed dreams, however. Over and over again, Freud himself shows that the day residue is associated with a significant and emotionally charged event of the previous day or recent days. In each case, the apparently innocuous residue appears to be substituted for the more significant experience during the process of dream construction because the significant experience is unacceptable to the dream censor.

    But what is the significance for the dream of the significant experience? Freud shows us how the dream censor acts to prevent unacceptable impulses from getting out of the Unconscious and into the dream. Here we see evidence that the censor is preventing the representation of an unaccept- able event in the store of newly acquired experiences from participating in the dream. It appears that the process of dream construction is facing in both directions at once, that it deals with inputs coming from both the Unconscious and the short-term memory structure that temporarily stores recent experience. Does the dream censor stand guard at the front door as well as the back?


    To answer this question more precisely, we will supplement Freud's introspections with a new kind of dream data, using a method of investiga- tion not available when Freud was writing. Freud's sample consisted of iso- lated dreams remembered and interpreted on the following day. We will look at a sequence of dreams, remembered and unremembered, inter- preted and uninterpreted, spread over several consecutive days and nights.

    This can be done if we have an analytic patient spend some of his nights in the sleep laboratory. He is awakened at the end of every period of dreaming sleep and asked to recount his dreams during that period. Then we match the dreams reported in the laboratory with the dreams reported during the analytic hours. We compare the associations to the dreams reported during the analytic hours with those of the dreams collected in the laboratory. (The findings described here are reported in detail in my earlier publications (5, 6, 7).) When we do this, we discover that dreams are not isolated events, propelled by unrelated impulses emerging, one at a time from the Unconscious. Instead, we find representations of particular recent and past experiences appearing and reappearing in new and differ- ent relationships on succeeding nights. We find direct evidence that when a significant day residue has been rejected by the dream censor on one night, it may be reincorporated into a new dream several nights later after a period of analytic interpretation and working 'through.

    On the first night the patient being studied had a dream in which a seem- ingly innocuous day residue had been substituted for an emotionally charged experience of that day. This innocuous day residue was superimposed on imagery derived from a highly charged memory of early childhood. The combination, as reported by the patient, made for a static and lifeless dream, producing few associations.

    During the patient's analytic hour, his resistance to the unacceptable day residue was worked through. On the night following this hour, in the sleep laboratory, he had a new dream in which the deleted day residue of the first dream was revived and superimposed in turn on the same memory of early childhood that appeared in the original censored dream. The new dream, which I call a correction dream, was filled with feeling and action. It sent out associative branches in many directions. In fact, it was from the associations to the correction dream that the original charged day residue and the early memory whose imagery appeared in both dreams were discovered.

    What we observe in this complex interaction is that the dream censor, operating through the mechanism of displacement or substitution, is engaged in a tug-of-war with an underlying adaptive process whose goal seems to be the condensation or superimposition of affectively significant experiences of recent and past events. What is this adaptive process, and what role does condensation play in it?

  • 306 PALOMBO

    When he describes condensation, Freud repeatedly refers to Galton's photographic method for determining family likenesses. He says the following (8):

    The dreamwork then proceeds just as Francis Galton did in constructing his fam- ily photographs. It superimposes, as it were, the different components upon one another. The common element in them stands out clearly in the composite pic- ture, while contradictory details more or less wipe one another out.

    Freud neglects to say that Galton's method was a test for congruence as well as a method for determining which elements might be responsible when a resemblance is actually present. He comes close to this idea when he remarks at one point in The Interpretation of Dreams (9), that "the pro- cess of unification into a single image may be said to have failed" when "the two representations are superimposed and produce something in the nature of a contest between the two visual images."

    Galton's photographic method produced an important negative result when applied to the search for a "criminal physiognomy." As Freud noted, the process of condensation in dreams may also produce a negative result. He does not notice, however, that this observation conflicts with his view that the dream can materialize only after a condensation has succeeded in adding the energies of its components and disguising their contents. This is important. If the process of condensation is a test for congruence between new experiences and old ones, then a number Of possible adaptive uses for such a function suggest themselves. Basic to most of these possible uses is the determination of an associative link to be formed in long-term memory between the new experience and the past experiences successfully matched with it.

    Support for this idea comes from the repeated demonstration in animal experiments that the learning of complex experimental tasks is consoli- dated during dreaming sleep {10). Deprivation of REM sleep following the learning of such a task will markedly diminish retention of the new learn- ing, as will destruction of the brainstem centers involved in the initiation of dreaming sleep,

    To summarize, dreaming appears to be an intermediate step in the over- all process through which new experience is assimilated into the structure of memory. Condensation is not primarily a defensive operation, as Freud suggested, but a test of congruence that determines which associative links will bind the new experience into the structure of the dreamer's past.

    But if the dreaming is a routine operation in the daily work schedule of the psychic apparatus, how are we to understand its sometimes rather spectacular achievements? Altman (11) asks, with understandable urgency,


    "whether the postulation of a matching machine in the service of adapta- tion adequately explains the richness of mental activity in its cognitive and affective aspects?" His reply is negative. "Even if the adaptive process proves to be a necessary concept, it is insufficient to account for the genius of the dream in producing Kekule's benzene ring, or inspiring Robert Louis Stevenson's literary imagination, Keats's poetry, Stravinsky's musical inspi- ration, Martin Luther King's social ideals, and, perhaps, the very source of psychoanalysis."

    The genius of the dream is certainly strange and admirable. But it is no stranger than other natural processes that recombine existing elements to produce new shapes and configurations. We experience the same mystery when we contemplate the generation of unique individuals through the matching process of sexual reproduction or the origin of new species through natural selection. The magic works because very complex ele- ments are being matched, and the mechanisms that bring them together are~not restricted in their experiments by narrowly logical categories (12). If the opportunity for new combinations arises repeatedly, as we now know it does in the regular cycles of dreaming sleep, then the emergence of novel forms and qualities becomes inevitable.

    The matching process in dreaming, in contrast to the secondary process organization of waking thought, is designed to produce novel and unex- pected juxtapositions of events ranging over the entire life experience of the dreamer. The dramatic action in A Midsummer Night's Dream provides an especially vivid example of this. Compulsive emotional states in two major characters are altered by the action of the play: Demetrius's love for Hermia and Titania's rebellion against Oberon. The transferential nature of these conditions is revealed as they suddenly dissolve at the moment of awakening from an enchanted dream (i.e., a correction dream).

    Demetrius actually awakens twice. At the first, early in the play, he merely exchanges his transference to Hermia for an equally irrational pas- sion for Helena. There is evidence neither of insight nor of working- through at this stage of the cure. The second awakening, late in tlhe fourth act, gives us an entirely new Demetrius. Something has clicked into place, a new connection formed by the matching of his infatuation with Hermia with his childhood fantasy of excited arousal, "the remembrance of an idle gaud/which in my childhood I did dote upon."

    Titania's disengagement from her infantile object is stated more succinctly: "Methought I was enamoured of an ass." The two cases are comple- mentary. With Demetrius the dramatic emphasis is on his present preoccu- pation, Hermia. The connection with his childhood fantasy is presented in a poetic image that recalls rather than recreates the childhood experience. Titania's transference love is not so clearly focused. At one moment it is

  • 308 PALOMBO

    oedipal, aroused by her jealousy over Theseus's impending marriage. At other times it is directed at the changeling boy, compulsively linked to the memory of his dead mother. The connection with the past is not explicitly verbalized, as it is with Demetrius, but in this case the childhood fantasy is recreated in action when Titania woos the monstrous Bottom, accom- panied by her imaginary companions, the delicately preoedipal fairies.

    As Shakespeare takes great pains to tell us here, the matching between present and past accomplished by the dream is neither simple nor mechani- cal, but an act of discovery capable of drawing on the full powers of the creative imagination. In the same way, Kekule's famous discovery clicked at the moment when a dream (one of how many?) succeeded in bringing together his present concern with the arrangement of the carbon atoms in the benzene molecule and the psychodynamically very primitive image of a snake biting its own tail.

    In order to understand this, we must overcome a very common miscon- ception about the nature and function of the human associative memory. The items we store, our "memories," are not neutral pictures of events, pas- sively filed in static containing structures. Each item of memory is a self- contained program of action for recreating the gratifying past event it represents in its imagery component. It is available to be activated when the present state of need resembles the earlier one.

    Moreover, the organization of the associative pathways leading to and from a particular stored program determines the set of circumstances under which that program can be called into action. This organization is a dynamic one, subject to continual updating and revision, and vulnerable to defensive interference at many points.

    Failures in such a system are as likely to be due to deficiencies in the associative structure of the memory as to deficits in actual experience. Kekule's achievement did not result from his acquisition of new facts but from his forming a new associative structure by matching old and new experiences already present in the various compartments of his memory.

    Psychoanalytic therapy can succeed only if it is capable of expanding and reordering the associative network in long-term memory (13). One of Freud's most important clinical contributions was his discovery that neu- rotic psychopathology is a misinterpretation of current reality brought about by the influence of distorted memories of childhood experience. Although these childhood memories have been rendered inaccessible to consciousness by the mechanisms of defense, their programs are never- theless repetitively called into action by stereotyped environmental signals.

    The process of "making the Unconscious conscious" must, at a mini- mum, restore the missing associative links that would ordinarily connect these repressed childhood memories with memories of later events whose


    action programs are under conscious control. Dream interpretation plays a special role in this process through its capacity to reengage the dreamer's adaptive mechanisms for bringing isolated memories of the past into real- istic alignment with his present life. This "realistic alignment" is not a pre- dictable response to an average expectable environment, but a new creation of the therapeutic process.

    Shakespeare does not give us a theoretical discussion of the process through which Demetrius is transfigured by his dream experience. Yet the sense of past and present mingled in a kind of double vision drifts through the lovers' banter after their awakening:




    These things seem small and undistinguishable, Like far-off mountains turned into clouds. Methinks I see these things with parted eye, When everything seems double.

    So methinks; And I have found Demetrius like a jewel, Mine own, and not mine own. (IV, I, 186-191)

    Demetrius's dream differs from those of our patients in that it offers little resistance to interpretation, like most of the dreams in Shakespeare's plays. (Bottom's dream that "hath no bottom" is a notable exception.) Shakespeare does not require a separate act of interpretation to intervene between Demetrius's experience of his dream and his subsequent understanding of it, even though the presence of Theseus is consistent with the therapeutic role he frequently plays. (The same is true of Oberon in relation to Titania.) The employment of the dream report as a dramatic device emphasizes the adap- tive function of dreams in drawing out the dreamer's underlying motives, rather than the defensive interference of the censorship mechanisms.

    It was Freud's much later discovery of the dream censorship that made us aware of the need for interpretation. Nevertheless, Freud's theory of dream construction tends to confuse the act of dreaming with the act of interpreting the dream. The fact that Freud was both the dreamer and the interpreter of his own dreams may have had something to do with this superimposition. His word for interpretation, Deutung, has the same root as Deutsch. In its earliest usage Deutung referred to the translation from official Church Latin into the everyday language of the Teutonic populace. In order to emphasize that dreams are meaningful, Freud gave them the special status of a sacred text and referred to interpretation as the transla- tion of this text into the "language" of waking thought.

    In this view, the interpreter can do nothing more than uncover the meaning already present in the content of the dream, however foreign the language in which it has been expressed. But in order to account for the

  • 310 PALOMBO

    disorderliness of the dream language, Freud had to assume that the text of the reported dream was itself a translation from everyday thought into a kind of secret code intended to deceive the censorship. The interpretation would then be understood as a retranslation from the dream code back into the original everyday meaning of the dream.

    But wherewas the original, everyday meaning to be found? If the mani- fest dream is a message in code, then the relationships among its compo- nents should mimic the relationships among the elements of the original message. But Freud insisted that this is not so, that the meaning of the dream is totally independent of the form in which it is actually experi- enced. The manifest dream now appears to be a decoy, rather than a trans- lation. The student of dream interpretation is advised not to be distracted by it.

    This puts the would-be interpreter in an odd position. Although the meaning of the dream is not to be sought in the manifest content, it must still be located within the confines of the dream itself. Otherwise the idea of a hidden instinctual message would lose its power. Since Freud had dis- covered that the hidden meaning could often be reconstructed from the dreamer's associations, he was obliged to consider these associations to be the actual, though "latent," contents of the dream.

    In this way Freud could maintain his view of the interpreter as the trans- lator of a disguised message. It follows that the interpreter can discover only what the dreamer already knew but managed to conceal during the con- struction of the dream. Just as the manifest cont~ent is a disguised version of the latent content, the interpretation is still another version of it, undisguised this time, and essentially identical with it.

    In the information-processing theory elaborated in my book, Dreaming and Memory (7), the dream and the interpretation are sharply distin- guished. The meaning of the dream is created by the dream in the carrying out of its function as a dream. An early memory concerned with the gratifi- cation of an infantile wish is necessarily a component of that meaning, but the meaning as such resides in the relationship established between the wishful infantile memory and the dreamer's current experience, as in the cases of Demetrius and Titania.

    An interpretation, on the other hand, reconstructs the relationship be- tween past and present components of the composite dream image and those other aspects of the dreamer's experience associated with them in the structure of his memory. The interpretation also extends the meaning created in the dream by integrating it into the more comprehensive con- ceptua ! structures available to the dreamer during waking life.

    Freud's own interpretations make it difficult to believe that as a dreamer he already knew everything that as an interpreter he was to discover.


    The complex and fascinating story woven by the wakin 8 Freud is a new construction, built on the cognitive outlines of the dream materials, but adding to them at higher levels of logical and narrative signification.

    In 1942, long before the discovery of rapid eye movement sleep, Samuel Lowy (14) had suggested that the dream functions at two distinct levels durin 8 sleep and in waking recall. He proposed that the primary function of dreaming is the creation of new connections during sleep. This function is normally accomplished without the retention of dream imagery in a form accessible to waking consciousness. Successful dreams are not ordinarily remembered. However, if the dreamer is awakened by the dream or dur- ing the dream, its contents will be remembered and may be put to work as a stimulus for more complex forms of conscious problem-solving activity.

    The findings of the sleep laboratory have confirmed Lowy`s interesting conjecture. In fact, the dreams that act as secondary stimuli in waking life appear to be precisely those dreams that fail in their primary function because the censorship mechanisms have interfered with them (7). When the dream censor is active, as it is in the small minority of dreams that alarm and awaken the dreamer, the result is usually a static and lifeless product that justifies Freud's characterization of dreaming as a "patho- logical mental activity" (15) analogous to neurotic symptom formation. But in the majority of dreams, which go unremembered and never come to the attention of either the dreamer or the analyst, the evidence of censorship activity is relatively minor.

    Freud's idea that a dream is the translation of a wish leads him into another very odd conception. He suggests repeatedly that the original form of expression of the wish is verbal and that the predominance of imagery in the dream results from a regression that serves to deceive the censorship by substituting a more primitive mode of representation. Here again Freud has confused the issue by reading the output of his interpretive activity back into the origin of the dream.

    Dream imagery is not a code, but the imagery of significant events asso- ciated in memory with action programs for reproducing those events. In order for the information content of the imagery to be accessible during the dream, the associated action programs must be prevented from going into action. This is not easy for the central nervous system to do. A massive inhibition of the entire voluntary muscular system, which occurs only dur- in 8 dreamin 8 sleep, is required to brin 8 it off. When brainstem inhibitory centers are ablated in the cat, normally patterned large muscle movements appear during REM sleep (16).

    When we interpret a dream, we make it possible for an undesirable action program to be intercepted and controlled by other means than the automatic but temporary inhibitory counteraction built into the dreamin 8

  • 312 PALOMBO

    state. The interpretation moves the action program beyond the narrow context of its original registration in memory and integrates it into a system of priorities more appropriate to current reality. In Demetrius's poetic lan- guage, it melts the transference into "the remembrance of an idle gaud." Some of the implications of an adaptive dream theory for therapeutic tech- nique are touched on in two recent papers of mine (17, 18). The topic is far from exhausted.

    In conclusion, Freud's theory remains a powerful and clinically useful statement about the psychopathology of dreaming, though no longer, I think, a reasonable explanation for the phenomenon of dreaming in its entirety. In our present state of knowledge it seems very unlikely that the censorship mechanisms work directly to oppose a random emergence of unacceptable impulses from the Unconscious. Instead, they interfere at various points in the complex adaptive process that links action programs in memory with current states of need and desire. The genius of the dream is a natural outgrowth of this complex adaptive process. Although it rises to the highest levels of creativity only rarely, it functions as the silent ally of the analyst throughout his day-to-day therapeutic work.


    In Act V of A Midsummer Night's Dream, Theseus and Hippolyta exchange views on the dreamlike adventures reported by the young lovers. Theseus dismisses their stories as fantasies of wish fulfillment, but Hippolyta points out that despite their strangeness, the tales reflect an adaptive change in the psychic reality of the lovers. The dramatic action of the play supports Hippolyta's view. The release of Demetrius from his transferential infatua- tion with Hermia comes at the moment of awakening from a dream in which he has matched his current feelings for Hermia with a repressed libidinal fantasy of childhood. This example of a correction dream illus- trates how condensation in dreams functions adaptively in matching a new experience with previously stored representations of related events in the past. It also illustrates the ability of the matching process to go beyond the narrow logical categories of waking thought to reach deeper levels of expe- rience otherwise inaccessible to the dreamer. This ability accounts for the important role played by dreaming in the creative process generally and in the day-to-day working-through process of psychoanalytic therapy.

    The adaptive function of dreaming is subject at many points to interfer- ence from the censorship mechanisms discovered and emphasized by Freud. A theory of dreaming combining these antagonistic processes is more consistent with the data of the sleep laboratory than the traditional psycho- analytic tl~eory alone. It also provides a better fit with the introspective data


    more familiar to the analyst as illustrated by Freud's well-documented analysis of his own dreams.


    1. Devereux, G. Dreams in Greek Tragedy. Berkeley: University of California Press, 1976.

    2. Brooks, H. F. (Ed.) The Arden Shakespeare: A Midsummer Night's Dream. New York: Methuen, 1979.

    3. Grinstein, A. Sigmund Freud's Dreams. New York: International Universities Press, 1980.

    4. Palombo, S. R. The dream and the memory cycle. Int. Rev. Psychoanal., 3: 65- 83, 1976.

    5. Palombo, S. R. Dreams, memory and the origins of thought. In J. H. Smith (Ed.), Psychiatry and the Humanities, Vol. 2. New Haven: Yale University Press, 1977, pp. 49-83.

    6. Palombo, S. R. The adaptive function of dreams. Psychoanal. Contemp. Thought, 1 : 443-476, 1978.

    7. Palombo, S. R. Dreaming and Memory: A New Information-Processing Model. New York: Basic Books, 1978.

    8. Freud, S. On dreams (1901). Standard Edition, Vol. 5. London: Hogarth, 1953, p. 650.

    9. Freud, S. The Interpretation of Dreams (1900). Standard Edition, Vols. 4-5. London: Hogarth, 1953, p. 324.

    10. Fishbein, W. (Ed.) Sleep, Dreams and Memory. Jamaica, N.Y.: Spectrum Publi- cations, 1981.

    11. Altman, L. Review of S. R. Palombo. Dreaming and Memory. Psychoanal. Q., 49: 320-323, 1980.

    12. Palombo, S. R. The cognitive act in dream construction. J. Am. Acad. Psycho- anal., 8: 186-201, 1980.

    13. Palombo, S.R. The associative memory tree. Psychoanal. Contemp. Sci., 2: 205-219, 1975.

    14. Lowy, S. Biological and Psychological Foundations of Dream Interpretation. London: Keegan Paul, Trench, Trubner, 1942.

    15. Freud, S. New Introductory Lectures (1932). Standard Edition, Vol. 22. London: Hogarth, 1964.

    16. Sastre, J. P., and Jouvet, M. Les schemes moteur du sommeil paradoxal. In W. P. Koella and P. Levin (Eds.), Sleep 1976. Basel: Karger, 1977, pp. 18-23.

    17. Palombo, S. R. How the dream works: the role of dreams in the healing process of psychotherapy. In S. Slipp (Ed.), Curative Factors in Dynamic Psychotherapy. New York: McGraw-Hill, 1982, pp. 223-242.

    18. Palombo, S.R. Deconstructing the manifest dream. J. Am. Psychoanal. Assn. (in press).

    Reprint requests to Stanley R. Palombo, M.D., 35 Wisconsin Circle, Chevy Chase, MD 20815.