We have a good idea today of how the body enables us to digest our food, or to run a mile in four minutes, but matters are very different when we turn to our mental life. Of course, we know that it is dependent in all sorts of intimate ways on the working of our body and in particular of our nervous system. When, for example, a man sees a physical object we have good reason to believe that certain events have taken place in certain parts of his brain. If he suffers from a head injury that affects the occipital area, we have good reason to expect that his vision will be disturbed in characteristic ways.

However, this growing accumulation of knowledge is still very restricted in its scope. Any item or collection of items from this accumulation tells us only about conditions in the body and nervous system that are necessary for us to see, hear, feel pain, remember things, and so on. It does not tell us what neural and bodily function is sufficient for these purposes. For example, in order for us to see, say, a chair in the way we do under normal conditions, it is necessary for the striate area in the occipital cortex to be present and functioning normally. But its presence and normal functioning are not themselves sufficient to explain the experience we have when we see a chair. Moreover, our knowledge of necessary conditions amounts to very little. While, therefore, we know enough about the digestive system to be able to give a very good account of how the body enables us to take in nourishment, we simply do not know anything like enough to be able to offer a good account of how the body and, in particular, the nervous system enable us to live our mental lives.

Nevertheless, the growth of our knowledge about the body and the nervous system, in spite of its severely restricted scope, has been so great and influential as to give rise to a widely accepted presumption both among the educated public and among scientists. This is the presumption that all our mental life can be explained as the manifestation “somehow” of the structure and functioning of the biological material making up the body and especially the nervous system. Of course, many scientists are not interested in this presumption. A psychologist working on cognition, or learning, or skills, or in social psychology can get along quite well without bothering about it. So can the many psychiatrists who are not concerned with the biological aspects of their subjects. But matters are very different with those in science who are concerned with the biology of human functioning. For, with them, this presumption is apt to regulate and even dominate their professional work and thought.

What about those specialists inside psychiatry known as psychoanalysts? How do they view this presumption? They seem to be as divided about it as they are about many other matters. To some of them, and I think they are a minority, Freud was just wrong in attempting (as he did in his early Project1 ) to explain the working of the mind in biological terms. In this attempt, his aim, he tells us, was “to represent psychical processes as quantitatively determined states of specifiable material particles,” which he assumes to be the neurons of the nervous system. In the view of the minority of analysts I have mentioned—of which Roy Schafer is probably the most prominent—Freud was equally wrong in trying to construct a general theory of a scientific character about his own discoveries, which he called “metapsychology.” In this psychology, mental phenomena are described and explained according to their location in the parts of the psychic apparatus; and by reference to the ways in which energy is distributed and attached to mental items inside the apparatus.

But, according to analysts such as Schafer, this attempt to construct a metapsychology is a fundamental mistake on Freud’s part. For, in essence, science is concerned with the causes of things, whereas psychoanalysis is concerned to uncover the reasons for a person’s feelings and conduct, and the meanings that the experiences of life hold for him. No scientific theory about energy and the parts of the psyche is logically capable of dealing with them.2

In sharp contrast with analysts who take this view, there are others who are close to biological and psychological inquiry, and who wish to see how far contemporary knowledge about the body and nervous system can be used to explain mental life as it is understood in psychoanalytic terms. In other words, they accept the regulative presumption I have sketched, and are interested in seeing what can be done today to carry through Freud’s ideas. Winson and Reiser both pursue this interest in their books.



Winson seeks to understand “how man’s mental life is derived from the physical functioning of the brain”; and he suggests that there is “a link between brain and psyche.” This is to be found in “a synthesis of neuroscience and Freud’s findings”—a synthesis which he proposes to offer us. Reiser opens his book in a similar vein by stating “a challenge: do psychoanalysis and neurobiology…have anything to say to each other?” “The immediate task,” he says in answering this challenge, is “to identify the translation rules that map” psychoanalysis into neurobiology.

These two projects sound exciting. But unhappily they run into a number of serious difficulties and objections. Winson concentrates on some of the central themes of psychoanalysis: the great importance of early childhood experience in the development of personality, the phenomena of the unconscious, repression, dream distortion, and transference. He claims that certain neural mechanisms and findings will provide “all the information we need to understand the biology underlying the psychological phenomena I have just described.” Thus he points to the neural mechanism underlying the importance of early childhood experience by drawing attention to, for example, the work of Blakemore and Cooper on the neuronal development in the visual system of kittens.3 Kittens were reared from two weeks to five months in the dark or, for some hours every day, in an environment in which all they could see were either horizontal or vertical lines. At the end of five months a kitten with experience of horizontal lines alone gave no evidence of being able to see vertical rods. Conversely for a kitten with experience of vertical lines alone. When the visual cortex of these animals was examined electro-physiologically, it was found that kittens exposed only to horizontal lines were deficient in units in the cortex that were sensitive to vertical lines. Conversely for the kittens exposed to vertical lines alone.

Winson argues that this work, along with much other research, supports the concept of critical periods in the organization of perception and learning, and so also supports the psychoanalytic view of the importance of early childhood experience. I find this argument very weak. It is well known that the psychoanalytic view of early experience is still the subject of inquiry and debate among psychologists. But even assuming that it is in general correct, we have next to no idea of how the brain of the human infant enables it to get through its supposed oral, anal, and Oedipal periods. All we can claim at the most is that Blakemore and Cooper’s findings are not inconsistent with the psychoanalytic view of critical periods and what we know about them. But this claim is a small one.

Winson spends much time telling us about the roles played by the parts of the brain known as the limbic system and the hippocampus. The former is a set of interconnected structures roughly bordering the brain stem; it is intimately connected with the cerebral cortex, and is generally agreed to be a center for emotional responses. The hippocampus is one of the limbic structures. Winson tells us that “the limbic system is the central core processor of the brain, in which memory and emotion are integrated, and the hippocampus is its gateway.” I strongly suspect that brain scientists are likely to regard this remark as a grossly exaggerated estimate of our present knowledge about these structures.4 But even if Winson’s inflated view of the matter is accepted, he fails to show how these findings are to be synthesized with Freud’s findings about the memories and emotions of his patients. Indeed, the only connection between them seems to be that, if brain scientists had failed so far to discover any neural basis for memory and emotion, it would be more difficult to accept the belief of many analysts, including Freud, that these phenomena have a neural basis. But the same difficulty would apply to any other psychology of memory and emotion, not just the psychoanalytic.

Winson does not appear to offer any neuroscientific findings connected with transference. On repression he refers us to a single complicated experiment, which he claims shows that the neural impulse in perception is subject to a delay, during which repression could occur. 5 Even if we grant Winson’s interpretation of this experiment (and I suspect that other interpretations are quite possible), we must remember that Freud’s concept of repression is a very complex one. All Winson has done is to show that one aspect of it—according to one view of how it works psychologically—may possibly be explained by the nervous system in the way he outlines.

Winson has much to say about the unconscious and dreams, and writes at length on comparative anatomy and psychology, as well as on rapid-eye-movement (REM) sleep, in animals and men. But I regret having to say that I found his argument here so confused that I am unable to present it briefly and fairly. However, in the light of the work he cites, he ends by presenting an account of the unconscious that, he claims, is “somewhat different from Freud’s.” Instead of the unconscious being a “cauldron of untamed passions and destructive instincts held in check by repression” he finds it to be a “cohesive, continually active mental structure which takes note of life’s experiences and reacts according to its own scheme of interpretation and responses. This reaction is reflected nightly in our dreams.” This view of the unconscious may well be supported by the evidence he cites from comparative anatomy and psychological work on REM sleep. But it is also quite evident that little of this evidence is drawn from the neurosciences. Hence he fails in his aim of synthesizing neuroscience with Freud’s findings about dreams and the unconscious.



Reiser describes at some length a psychoanalytic case of his own, which lasted some years. This was a case in which the patient recovered a large number and variety of memories, which formed a complex and enduring memory network. This network centered around an early and traumatic experience—the sudden death of the patient’s mother while giving birth to her brother—which then developed associations with “emotional analogues” of the early experience in later life. These associated events in turn became centers of emotional disturbance for the patient. Reiser sketches the structure of this memory network, involving the early and later experiences, in several “cartoons” or drawings. But Reiser does not try to offer any rules to enable us to match these memories, uncovered in psychoanalysis, with the mechanisms of neurobiology.

When he turns to neurobiology, he asks, “Where in the brain are memories stored?” and “How may nerve cells remember?” In answering the former question, he refers with some admiration, and I believe justifiable admiration, to some current work in the field. But he concedes, in effect, that we do not have an answer to the question about “where.” He answers the “how” question with an “approximation” answer by referring to the work of E.R. Kandel and others on the marine snail Aplysia.6 These researchers conditioned a snail to take defensive action against shock when it encountered an inoffensive or neutral stimulus. This is known as “aversive” conditioning. They then found evidence strongly suggesting that certain biochemical changes occur which affect the synapses of the specific neurons concerned in this learning; and that these changes explain how the snail remembers that the inoffensive stimulus is a signal of shock. However, I am sure that Reiser would agree that the admirable work of Kandel and others is still in its early stages; and that we have not yet come near to explaining the basis either of human memory or of the memory networks in which he is interested.

Reiser is primarily concerned, however, to make Freud’s developed theory of anxiety scientifically respectable and to “map” it into neurobiology. I think he is wise to try to do this, because of the key role that Freud’s view of anxiety plays in psychoanalytic theory as a whole.7 On this view a person (for example, a young child) experiences anxiety in a situation of danger and helplessness. If the situation threatens to recur in later life, the person experiences anxiety as a signal of impending danger. This enables the person to take anticipatory defensive action, and so homeostatically to restore and maintain psychic equilibrium.

Reiser now offers us “a detailed recasting” of Freud’s view; and shows in some detail how this view can be restated according to Pavlov’s model of conditioning. I think this theoretical exercise is helpful and clarifying; and, as he says, it will make Freud’s view “easier to think about in a broader biological context and to study both in the clinic and laboratory.” But such theorizing does not in itself do much, if anything, to help to map Freud’s theory into neurobiology. For the Pavlovian model stands in equal need of neurobiological explanation.

When Reiser moves courageously into neurobiology, he makes several suggestions in the light of current work which take us into very technical details of anatomy and biochemical functioning. In particular, he picks out a small structure in the brain stem called the locus ceruleus (LC for short), which we know contains and releases norepinephrine (NE), a substance which we have good reason to believe is connected with the production and control of anxiety. The evidence also suggests that this small nucleus may serve by releasing NE to “alarm” the person, and thereby to produce signal anxiety. In addition, the association cortex and the limbic system serve to enable the person to learn and store information about the situations that have threatened him in the past; and these parts of the brain have connections with the LC. So, to put it crudely, if some current situation is like a past threatening one, these parts of the brain somehow treat the present situation as being like the past one, and so give it this “meaning” to the person. The parts then activate the LC, which sets going feelings of signal anxiety and defensive action by the person.

I believe that the suggestions Reiser discusses here will prove interesting and stimulating to workers in this field, but our relevant knowledge is still elementary and incomplete; and his neurobiological account of our anxiety reactions strikes me as being a very speculative one. I think Reiser still has a long way to go on the lines he discusses before he will be able to persuade critics that the anxiety a person feels can be brought within the scope of science. In the later part of his book Reiser offers us a number of interesting thoughts on the psychophysiology of stress. But he leaves us very unclear about just what the connection is between these thoughts and psycho-analysis.


The schemes proposed by Winson and Reiser sound very exciting, but it is evident that the authors have not been successful in carrying them out. Winson has not given us a synthesis of neuroscience and Freud’s findings. Reiser has not given us any “rules” that “map” psychoanalysis into neurobiology. Their efforts to substantiate their claims must be regarded as failures.

What has gone wrong? I think the chief source of the trouble is to be found in their view of the present state of neurobiology. It is quite true that there have been enormous advances in our knowledge of the brain and of the nervous system during the last few decades; and this knowledge is still expanding rapidly in the various branches of psychobiology. But it is very misleading of Winson to conclude his section on “neural mechanisms” by saying, “All of the evidence is now in: We have seen the machinery of the brain that deals with perception, memory, emotion, sleep, and early development, and the psychological phenomena, that is, the unconscious and its various ramifications.” Similarly it is very misleading of Reiser to suggest we should look for “translation rules,” when he also concedes that we do not really know how human beings remember and where they store their memories. For this means that we still do not know where to start to do the translating. It seems clear, therefore, that Winson and Reiser overvalue the findings in neurobiology, and exaggerate the utility, for their purposes, of current scientific knowledge.

What also helps to lead the authors into trouble, I think, is their concentration on psychoanalysis as the psychology they seek to link with neurobiology. Because psychoanalysis contains a theory on a grand scale, they imagine that neurobiology is able to produce connections with psychoanalysis on an equally grand scale. They ignore that the neurobiology of the brain is still in an exploratory stage, able at best to draw attention to some states and processes in the nervous system that are, or may be, necessary for one mental function or another. Because of this, the attempt by the authors to derive mind from body is bound to end in fragmentation and failure—as happens in these books. The authors do not get far in trying to carry through Freud’s early ideas. The ambitious strategy they propose raises high hopes, which are then disappointed. If Winson protests that he was really only concerned to put forward a hypothesis about the brain mechanisms that underlie the unconscious, then his hypothesis has very feeble support. If Reiser protests that he was really only trying to answer the question: “Do psychoanalysis and neurobiology have anything to say to each other?” then the answer is: “Very little.” All this is natural enough; for their entire enterprise is premature.


But why bother to look at the matter through psychoanalytic spectacles at all? If our overriding aim is to find the connections between body and mind, and to try to see what neurobiology can do to explain mental functioning, why restrict ourselves to a psychoanalytic view, or a psychoanalytically influenced view, of the mind?

If psychoanalysis offered us a theory of mental functioning that has been established, then it would be not only reasonable but also necessary for our authors to look at the mind through psychoanalytic spectacles. But analytic theory has not been established. On the contrary, it is embedded in controversy, its epistemological status is still unclear,8 and it is very far from being generally accepted in psychiatry itself.9 Our two authors do not give anything like sufficient weight to the limitations of the theory, which generate the controversy and uncertainty about it. Thus they do not draw adequate attention to the central problem of the theory, which is whether it can be supported or falsified in the same way as a scientific theory; and, if not, how it is to be supported or falsified.10 Indeed, I can hear some critics accusing them (and especially Winson) of being victims of the American myth that psychoanalysis is a great psychology which has shown us, in essentials, how the mind works. I suspect that their naive attitude to the status of psychoanalysis has helped to mislead Winson and Reiser into pressing the connection between neurobiological inquiries and the psycho-analytic view of the mind.

Psycho- and neurobiologists would generally agree, I think, that if one is interested in showing how the nervous system can explain our mental functioning, it is wise to be modest. If one is interested in aversive conditioning, then by all means study the biochemical changes involving specific neurons—changes that seem to be necessary to enable the snail Aplysia to remember that an inoffensive stimulus is a signal of shock, and so to take defensive action. This is a study that Kandel and his associates have pursued with important results. But do not mislead yourself and others into supposing that you are talking here about psychoanalytic metapsychology or anything close to it. For we have no reason to suppose that, when the snail takes defensive action against the inoffensive stimulus, it is experiencing any feeling at all; a fortiori, we have no evidence that it experiences what a person does when, according to Freud, he feels signal anxiety. Try instead to extract something of value from the struggles of psychologists to deal with different types of memory—such as short- and longterm memories; memory of episodes; and memory involving the meaning of words and concepts.11

We can also look with profit at the slow piecemeal work—to which Winson rightly draws attention—going on in neurobiology to discover how different parts of the brain are necessary for different types of memory.12 Obviously, it is exciting to find the single neuron in the monkey that appears necessary to enable the animal, when presented with a set of objects one at a time, to give a recognition response to an object in the set which had been previously presented.13 But we have to be careful not to inflate this discovery into supposing that we are now on the verge of discovering the single neuron that is necessary for conscious recognition and perception in human beings. 14

Moreover, it is important to bear in mind that, if one is interested in making some part of psychoanalytic theory easier to grasp and scientifically more acceptable, one does not have to wait for the development of neurobiology. There is something else that one can do now. One can try to clarify that part of the theory and then show how it can be brought within the scope of a scientific psychology. For I think Roy Schafer is right to protest that Freud’s theory of signal anxiety and ego functioning does not show how the reasons that influence a person and the meanings he assigns to events can be brought within the causal machinery of a scientific account. This part of Freud’s own theory of the mind is, indeed, very obscure and confused.

In trying to deal with this problem, Reiser would be better advised, in my view, to put neurobiology almost wholly to one side at this stage, and to concentrate instead on developing still further his efforts to clarify conceptually Freud’s theory of signal anxiety. He might do this by spelling out how the homeostatic mechanisms that are obviously required by the theory actually work. With this in mind, he could invite his colleagues in theoretical psychology to help him to develop testable models of the psychological machinery and functioning postulated by Freud’s theory. This is no easy task. But it would help to show that, however right Schafer, his one-time colleague at Yale, may be in protesting at Freud’s confusion, he is wrong in maintaining that Freud’s metapsychology cannot logically be incorporated into science.

All this is not to say that the two books under review are without interest. Far from it. They draw the attention of the ordinary reader to scientific findings about the body that may take us along the road to an explanation of mental functioning. Both, especially Reiser’s book, draw attention to findings that help to clarify some of the confusion surrounding the psychoanalytic view of the mind; and thereby they open the eyes of the skeptic to ways in which analytic theory may be brought within the scope of scientific inquiry. They present specialists with a number of speculations and suggestions that may turn out to be helpful.

This Issue

July 18, 1985