Adaptation of an article by Donald Nathanson, Executive Director of the Silvan S. Tomkins Institute, Philadelphia, Pa. Introduces the affect theory of Silvan Tomkins and its relation to conferencing.

(Reprinted from The Annual of Psychoanalysis, Volume 25,
edited by Jerome A. Winer for the Chicago Institute for Psychoanalysis.
©1997 by the Chicago Institute for Psychoanalysis.)

In an issue of this annual devoted to the memory of Michael Franz Basch, it seems only reasonable to show how one of his contributions has helped expand the power and depth of interpretation while providing new therapeutic force for our field. Although trained as a classical psychoanalyst and brought up within the ethos of drive theory, Basch recognized immediately the validity of Kohut’s (1971) observation that through some form of empathic attunement the mothering caregiver becomes aware of the infant’s inner experience and uses this awareness to guide all attempts toward soothing the infant’s roiling affects. The very existence of empathy strains drive theory to its core, and Basch took the phenomenology, mechanics, intrapsychic ramifications, and interpersonal aspects of empathy as his special pursuit.

In Group Psychology and the Analysis of the Ego, Freud (1921) acknowledged the common observation that wordless fear could mutualize throughout a herd or community of animals with great simplicity. Sullivan (1954) followed the lead of Scheler (1912) in calling this "contagion of emotion," but, like Freud, offered no explanation for the casual case with which emotion seemed to ignore the dielectric that insulates nearly everybody from the thoughts of others. Sullivan’s interpersonal psychology depended on such transfer of emotion-related information, whereas Freud’s concentration on the intrapsychic world allowed little attention to what might be going on in the "mind" of another. With his observation that self-object function included rapt attention to the inner world of the infant, the ability to mirror and thus experience the affective portion of that infantile experience, and the skill to modulate this raw outpouring of affect, Kohut unwittingly forced the development of bridges between those who see the individual as solitary within a crowd and those who see individual identity as the result of differentiation within a group matrix. With the wry humor that characterized so much of his work, Basch took the position that the interpersonal transmission of emotion required sensible and specific attention simply because it existed.

Contrast the two most extreme forms of emotion transmission—the subtle inferences drawn by the skilled psychoanalyst or novelist about the inner world of a partner in conversation as opposed to the wave of bawling that travels through a newborn nursery when any one of its denizens begins to wail. Primitive empathy, said Basch (1976), involves the transmission of the raw data of emotion as seen in the herd of newborns, whereas mature empathy requires that sophisticated adults experience such transmission and use their personal memory bank of emotional experience to figure out what might have produced that particular outpouring of emotion. Mature empathy is an outgrowth of the primitive transmission of emotion between individuals, even though this dichotomization provides no clue about the way such transmission occurs. Selfobject function may then be viewed as a subset of mature empathy, one that can be graded in its effectiveness on the basis of the levels of skill used to modulate the emotions of the sending organism.

Affect Theory

Finding nothing within psychoanalytic experience or theory to explain this realm of data, Basch now began to evaluate theories of emotion that derived from other sources. The only system of thought that provided answers for all his questions was the extraordinarily challenging and intellectually demanding affect theory of Silvan Tomkins (1962, 1963), which had been published and ignored a decade earlier. In brief, Tomkins stated that humans are wired or organized with three primary systems for the management of nonverbal information: the pain system, the drive system, and the affect system.

According to Tomkins (1962), the most primitive of the three systems involves pain, which is both localizing and motivating. What we experience as pain is not tissue damage as such, but a report about that damage made by specialized pain receptors within but unrelated to the actual tissue that has been injured and which therefore transmit the sensation of pain as an analogue of whatever injury has occurred. It is the report that motivates us by drawing our attention to the site of the injury so we may do something to limit the damage. This evolved separation between the bodily systems that handle injury and those that make reports about it allows such disorders as referred pain and chronic pain syndromes, in which little or no tissue damage can be found in the area that hurts (Nathanson, 1995a).

The drives, said Tomkins, are also a system of reports, and they have the following characteristics: 1) they call attention to the fact that some body function needs to be performed (usually transport of a substance into or out of the body as in hunger, the urge to eliminate, or sex); 2) tell us where consummatory action is to take place (we are hungry only in the mouth, aroused primarily in the genitals, tired in the eyes that must be closed in order for us to sleep); and 3) they initiate the process of consumption (sucking movements, hand to genital, closure of eyelids). The drive tells an organism what must be done and how to do it when that organism is incapable of making the conscious decision to do so. Drives provide precise localization of need and its satisfaction, but are incapable of providing the motivation to make it happen. We can ignore hunger, sexual arousal, or the need to excrete when something else seems for the moment more important; sexual arousal is a paper tiger when joined with any affect other than excitement or anger.

Just as pain is not tissue damage as such but a report about that damage, affect is neither the stimulus itself nor our appraisal of it, but a report about that stimulus based on the way it is received by the neural apparatus. Our connection to the external and internal world is partly the report of our sense organs and partly our affective response to those and other reports. The mechanisms for the acquisition of data are different from the affect system that assesses such matters as the rate at which this information enters the system. It is for this reason that we can have clinical syndromes that involve disorders of affect (distortions of the reporting mechanism) that are not caused by the information or data of the sort that is the normal trigger for affect but from malfunction of the physiological systems that make up the affect system (Nathanson, 1994a). Psychological events alone may indeed lead someone to become excited enough to behave in a manner that might be characterized as hypomanic, although the forms of mania seen in bipolar affective illness are caused by a quite separate mechanism.

It is on the affect system that the organism must rely to provide motivation. The affects are a group of innate mechanisms that respond to certain specific stimulus conditions by contracting and relaxing the muscles of the face into specific patterns of expression, and altering its microcirculation to produce redness or pallor. Associated with these facial patterns are certain vocalizations, alterations of pulse and respiration, and variations in endocrine and exocrine secretion. The newborn child is unable to appraise the world as a vale of tears—it cries because sonic triggering stimulus (like cold, hunger, fatigue, loneliness, or mild pain) is both noxious and steady-state. In its noxiousness and steadiness, the sobbing response is analogous to whatever has triggered it. Since the affect of distress now amplifies whatever steady-state stimulus had called it forth by securing the attention of the organism and its mothering caregiver to that steady-state stimulus, we say that the affect acts as an analogic amplifier of its stimulus conditions. Significant moments before the cry was vocalized one might have noticed that the baby’s eyebrows were arched and its mouth turned down in the pattern known as the omega of melancholy; high-speed photography (Tomkins, 1962) reveals that the facial expressions themselves linger over a timespan that is also an analogue of the characteristic steady-state profile of their trigger.

Although it might seem more logical for us to assume that the individual cries in response to its cognitive appraisal that some situation is noxious, a host of studies confirm the position taken by Tomkins. Expressions of distress are visible in situations that a newborn could not possibly "understand" as distressing. As one comes to understand the affect system it becomes clear that the face leads in its response to emotion-related stimuli. Nothing can be considered an emotion until something happens inside the brain, goes outside the brain to cause a specific pattern of facial expression that is read as such by kinesthetic sensors, and returned to the brain as an affective experience. The face is the display board of the affect system. Patterns of stimulus acquisition summarized by Tomkins (1962) as stimulus increase, stimulus decrease, and stimulus level are responsible for the six basic innate affects. As Demos (1983) has said, the affect system converts a quantitative signal into a qualitative experience. The affects operate like a bank of spotlights, each of which turns on and off in accordance with its innate physiological trigger, shifting whatever has triggered that affect from ground to figure. It is affect that focuses our attention on its trigger; nothing can be the subject of our attention unless and until it has triggered affect. Consciousness is a special channel of neocortical activity made available under the influence of affect. The innate affects provide the mechanisms through which a physiological stimulus becomes a psychological entity—everything we know as psychology is so because an affect has been triggered.

Tomkins developed a definitional system within which each of these six affects is given a group name, the first indicating the mildest presentation of the affect and the second representing its most intense presentation. What I have until now described as distress is more properly known as the range of distress to anguish (distress-anguish), covering affective responses from the mildest of sobbing through the keening wail of an anguished mourner. It is not that humans decide or choose the degree and intensity of distress to be exhibited on the basis of the importance of the distressing trigger, but rather that we learn through experience to gauge the gravity of a situation from the intensity and duration of our own affective response to it. As we travel from infancy through maturity, each experience is stored in the basket of analogous or similar affective reactions. We come to know life, our own individual lives, as sequences of stimuli, triggered affects, and our immediate responses to the initiating situation as defined by the affect associated with it. Life is not a matter of stimulus-response pairs (Mowrer, 1939) but of stimulus-affect-response sequences, simply because in an advanced organism no stimulus can possibly evoke a response until and unless it first triggers an affect.

There are, then, six of these basic affects: along with distress-anguish, the response to a low-grade, noxious, steady-state stimulus, are responses in the range anger-rage, the group of affects triggered when a stimulus is steady, noxious, and even higher in density than that required to trigger distress-anguish. Anger-rage is characterized by the contraction of muscles all over the body but especially those of the jaw, by the yell or roar of rage, and by reddening of the face that makes anger a hot affect. There exist also the gradient affects, built-in responses to stimuli that rise or fall in density: Any time a stimulus decreases in intensity we are wired to respond with pleasure. When the amount of decrease per unit time is gradual, the affective response involves a relaxed facies and smile of contentment or enjoyment, whereas a rapid decrease in stimulus density (as when a joke is terminated by its punch line) triggers the laughter of joy. Although Tomkins (1962) gave this affect the formal name of enjoyment-joy, I have tended to describe it as "contentment" to reduce the obvious but incorrect confusion of enjoyment with excitement (Nathanson, 1992). Any sudden loud noise (pistol shot, thunder clap, something dropped in the quiet night) bears the stimulus profile of a square wave (sudden onset, sudden offset) and triggers the range of affective response called surprise-startle in which the eyes open wide and blink, the eyebrows rise quickly, and there is a brief vocalization of "Oh!" while the mouth indeed resembles the letter O. Lasting only a few hundredths of a second, the brevity of the affect mirrors its temporal profile; surprise-startle only readies us for the next stimulus to come and may be thought of as a "reset button" for the affect system.

Two more ranges of innate affective responses to alterations in stimulus gradient remain: When data of any sort travel through any part of the central nervous system (perception, memory, cognition, information from a drive, events occurring during a dream), and those data enter the system at an optimal rate of increase (as when we hear something faintly and walk toward it to hear better), an affect is triggered over the range from interest through excitement. Interest-excitement is the affective response to novelty, to the acquisition of information that occurs at an optimal gradient. At the mildest end of its range, the affect furrows the brow, turns the head in the position we call "track, look, listen," and may also open the mouth slightly; this is usually accompanied by a mild and pleasant increase in heart rate and respiration. Full excitement is registered by an even more vigorous display of the same expressions, often with the mouth opened wide, the eyes bright and dancing. The affective response of interest-excitement is identical whether the stimulus comes from the steadily increasing goad of a mathematics problem, the act of composing a poem or painting, or sexual arousal.

When data travel through the system at a rate too rapid to trigger interest-excitement but not rapidly enough to bring out the response of surprise-startle, our response to too much, too fast, is a blanching of the cheek, the shift from "track, look, listen" to a stare, momentary freezing of the limbs, and a quite uncomfortable increase in heart rate. All of these stigmata of fear-terror become more intense at the upper end of its range, until terror itself may be experienced more as a problem with the heart than as a mere emotion. It has been my experience that those who present to emergency rooms with cardiac neurosis are unlikely to recognize the lower end of this affect’s range, and only "notice" it when pounding heart and drenching sweat amplify terror to such a degree that these sites of action for the affect protocol themselves become the focus of attention.

Of the core group of innate affects, two (interest-excitement and enjoyment-joy) are pleasant, positive experiences that we are wired to enjoy, and three of the remaining four are noxious experiences that we are wired to dislike in varying degrees. There are three additional mechanisms that have evolved into the status of negative affects: 1) a response to bad-smelling food for which Tomkins coined the name dissmell; 2) a response to bad-tasting food he called disgust; and 3) a response to positive affect that has been impeded but not stopped entirely, which he called shame-humiliation. Elsewhere, I have written (Nathanson, 1987, 1992) at length how this final physiological mechanism that acts as an analogic amplifier of an impediment to positive affect comes to fuse in development with the two aforementioned drive auxiliary mechanisms that protect us from hunger gone wrong, so to evolve into the complex and powerful psychology of the adult shame experience. Mature shame emotion is a mixture of shame affect (which interferes with the experience of both interest-excitement and enjoyment-joy) with self-dissmell and self-disgust. The complex world of adult emotion involves both the innate affects and their innate triggers, as well as our remembered storehouse of affective experience.

Tomkins (1981) noted often that through development the crisp edges of neonatal innate affect blur to the more muted and complex display of adult affect, just as young mountains are sharp and craggy, and older ones smooth. Even superficial perusal of Tomkins’s work forces the clinician to pay far more attention to the specific affects being expressed by an individual, for what is involved in emotional expression is not the meaningless energy released when a drive has been frustrated from reaching its aim, but an amplified analogue of a quantifiable stimulus of far greater significance. "Conflict" is only a minor example of the immense range of situations that can trigger affect, and much time in therapy is wasted looking for the putative conflict to which the affect of the moment can be assigned. Whereas the classically trained psychoanalyst tended to tap somewhat impatiently waiting for an expressed emotion to go away because it was only drive energy that would not have been wasted were the analysand to repair the damaged drive management system, one trained in affect theory studies each and every expression of affect in order to figure out the specific stimulus-affect-response sequence involved.


This was the complex language that Basch (1976, 1983) absorbed on his way to a new understanding of empathy. If an affect was an analogic amplifier of its stimulus conditions, said Tomkins (1962, 1981), then the facial and vocal expression of an affect broadcasts that affect both to the observer/listener and to the self of the organism. Each affect tended to trigger more of itself through a process of internal contagion (simply because the affect is a competent analogue of its originating stimulus), and was capable of triggering the same affect in another person who listened to or watched an individual who was experiencing an innate affect. Here was the physiology that underlay the affective resonance Basch described as "primitive empathy." Even though Tomkins (personal communication) preferred to view adult affective experience as nothing more than what happens when the innate mechanism is triggered within a highly complex matrix of nested and interacting ideo-affective formations, Basch suggested that we use the word affect to refer to the physiological expression itself; the old word feeling to represent what happens when we become aware that an affect has been triggered, and emotion to represent the combination of an affect with our memory of previous experiences of that affect. Using his logic, I have suggested (Nathanson, 1987, 1992) that affect is biology, whereas emotion is biography. You and I have the same nine innate affects—the same mechanisms reside in our subcortical brains—but our life experiences have been so different that only as we learn each other’s history can we know what either of us means when using an emotion label.

Now Basch had the language to explain the difference between primitive and mature empathy. If it is the association of a group of memories with the affect of the moment that produces the formal structure we call an emotion, then might not a similar process take place to produce mature empathy? When you and I experience an affect, we scroll through our own personal memories to find stimulus-affect-response sequences similar enough to match the affect of the moment. So much for normal emotion within each of us as individuals—but what happens when we are the receivers for affect triggered by another person who may therefore be called a broadcaster? Scenes found when the receiver scrolls through memory for a stimulus-affect-response sequence need not be of the actual experience that triggered this affect of the moment in the person whose broadcast we have received. Our ability to associate to this sequence consisting of affect resonated from another person, followed by our memory of what might have produced that affect had it been stimulated in us by the normal innate pathways and triggers, has allowed us to determine what might have happened to us to produce such a sequence. This is the basis of mature empathy—our ability to figure out a stimulus-affect-response sequence that might have occurred in another person and thereby to know a great deal more about the inner experience of that other. The success of our empathic attunement, the accuracy with which we have tuned in to the world of another person, is dependent on the degree to which we have allowed ourselves to resonate with the affect broadcast by that person, and on the quality of fit between our life experience and their own.

The Empathic Wall

In these 1976 and 1983 papers, Basch developed the logic that allowed us to understand how information about the inner world of the infant moved from infant to caregiver. In doing so, he introduced into the mainstream of psychoanalysis, clinical psychology, and psychiatry the system of thought that had been developed earlier by Tomkins but remained until then inaccessible. By chance, I had begun my own investigations into the nature of emotion transfer about the time of that second paper, to which I was steered by a senior colleague. My own interest was almost the exact reciprocal of Basch’s, because I took it for granted that the emotion of one person could infect or influence the emotion of another. What puzzled me was why and how people could become immune to the affective experiences of others. Why didn’t we live in almost constant interaffectivity, in the confusing world of generalized civic mutualization of affect? Basch’s use of Tomkins’s work seemed so reasonable that I adopted it whole cloth and started a study group so many of us could learn it at once. As I developed facility with this demanding language, I wrote both Basch and Tomkins, asking for assistance as I tried to use these new concepts in a sentence. These relationships, which became the center of my own intellectual life, lasted until their deaths in 1996 and 1991 respectively.

Basch (1995) took for granted that psychotherapy is applied developmental theory and based all of his suggestions for treatment on what we knew about psychological development. Tomkins saw the individual as the product of simultaneous development of many interlocking systems, including but not limited to the drives, affects, cognitive apparatus, motor, and sensory systems. I am interested in anything that works to change people, which forced me to include the study of psychopharmacology and all other somatic treatments. Why do some kinds of "depression" improve with the tricyclic antidepressants, others only with the selective serotonin reuptake inhibitors, and still others with benzodiazepines more traditionally associated with the relief of "anxiety"? Study of the individual affects led me to the conclusion that each pharmacological agent worked on the physiology of one or another of the nine innate affects, and that "depression" was not a disease but rather a group noun that referenced the mental state associated with the persistence of any one or more of the six negative affects (Nathanson, 1992; Nathanson and Pfrommer, 1996). Where Freud (1917) might say that depression was "anger turned inward," I replied that what he saw as "hostility" against the self was no more than the operation of shame affect associated with self-dissmell and self-disgust to form mature shame emotion, or the fusion product of shame affect and fear of reprisal that we know as guilt (Nathanson, 1987). It is shame that leads us to say "I am defective, disgusting, ugly, unworthy of love, a failure," and it is guilt that produces thoughts and statements like "I have done terrible things for which I will be punished." Over and over, we taught young clinicians that "you can’t diagnose depression in a patient who doesn’t make you feel depressed," thereby teaching but not explaining that affects mutualize and cognitive constructs do not.

Even more interesting was the observation that affects do not seem to mutualize equally for everybody. Directly proportional to the intensity of obsessional traits in a personality is the degree of opacity to the feelings of others, while those shame-ridden people labeled as "borderline" by those who pay little attention to affect (Nathanson, 1994b) are so preternaturally sensitive to the affective broadcast of others that often we spend a great deal of time in therapy teaching them how to distinguish their own emotions from those of others around them. Even among psychotherapists we see a wide range of empathic sensitivity, suggesting at the very least that empathy is neither something to be learned from texts nor a right conferred with the award of a degree or a license.

I know of no study suggesting that neonates differ widely in their sensitivity to affect broadcast into their environment; distress-anguish seems to spread rapidly and fairly evenly throughout a newborn nursery. (A prospective study of this sort, linking even "minor" variations in such sensitivity to later psychopathology, might be of great importance to our field.) Although the audience for a comedian or a demagogue is self-selected and therefore possibly more sensitive than the population at large to the affect broadcast by its chosen source, both positive and negative affect are mutualized there to a far greater extent than in normal interpersonal life. An audience is called hostile, tough, or difficult not only when it is antipathetic toward or even visibly angry at the work of its star attraction, but also when it is uninvolved with, unreactive to, or out of synch with its supposed leader. Lovers demonstrate an exquisite sensitivity to each other’s moods, a sensitivity that is relinquished rapidly when the relationship becomes troubled. Affective mutualization is easy, uncomplicated, and free-form in infancy, and subject to wide variation in adult life.

It seemed only reasonable, therefore, to suggest that free-floating affective resonance is the norm for early childhood, blocked at some point in normal development, and reestablished by those adults to whom it seems either interesting or unavoidable. I suggested that this block to primitive empathy was learned (Nathanson, 1986, 1989), offered as its label the "empathic wall," and demonstrated that it was essential for the formation of an adult personality. An adult who walked through life always vulnerable to the affect being broadcast into the local environment would be unable to maintain personal boundaries, just as an adult who admits no information from the affect broadcast by others is truly isolated. The empathic wall must be strong when necessary but possess doors and windows that can be opened when necessary and optimal.

How might this learned mechanism be constructed? It is axiomatic that every culture on the planet requires that children mute their display of affect by the time they are three years old; we teach this every time we "shush" a child and every time we value verbal over affective communication. To socialize a child is to teach it how to modulate the display of innate affect; socialization demands that a child yield its ability to take over interpersonal space through the broadcast of each affect at the maximal end of its range. Shaming labels like "immaturity," "childish," and "infantile" drive home this message about affective broadcast.

Spitz (1965) likened affective resonance to telepathy and gave it the name "coenesthetic communication." Concerned about the resemblance of interaffectivity to telepathy, in 1984 I visited the massive library in which several hundred thousand "psychic readings" of Edgar Cayce are housed and was unable to find a single "reading" that involved emotion rather than facts about the life of the individual being read. I have taken care to study every example of telepathy presented in scientific or lay literature and found no instance in which an adult with significant gifts in the telepathic acquisition of information seemed attentive to or involved with data about affect. Earlier, however, Freud (1921) had suggested that affective transmission is accomplished by a subtle mimesis involving an automatic imitation of the bodily position and facial expression of the other, and this seemed to fit better our current ideas about the nature of human emotion. Telepathy and affective resonance seem to be two separate channels of data acquisition, and it is to the latter that I have devoted my attention.

Freud (1921) and Tomkins (1962, 1981) agree that interaffectivity requires the receiver to mimic the affect display of the broadcaster, an action that (in Tomkins’s language) forces the receiver to experience an analogue of the affect going on in the broadcaster. I have suggested (Nathanson, 1986, 1989, 1992) that the receiver portion of the empathic wall involves the skill of damping or refusing to perform this mimesis, thus diminishing the degree to which the receiver can possibly be affected by the transmission of the broadcaster. The maintenance of facial stiffness in the presence of another’s affect display guarantees reduction in intrusive or unwanted primitive empathy and whatever benefit might accrue from knowledge of the inner world of that other.

There is a sexual analogue for the other moiety of the empathic wall, one known intuitively more by men than women. Sexual intercourse as such is terminated when the male partner ejaculates, since tumescence cannot occur again until a biologically determined refractory period has elapsed. Sexual activity is a prime example of an interaction made possible by the action of a drive, its intrapsychic amplification by the positive affect of interest-excitement, and its magnification into the interpersonal sphere by interaffectivity. In heterosexual intercourse, a man is likely to disappoint his sexual partner if he terminates the experience before that partner has achieved orgasm; thus, sexual competence requires strategies for controlled immunity to the broadcast of his partner that will lead to premature ejaculation unless subject to internal modulation. For most of us, this is accomplished by shifting attention to some other source—as when we think momentarily about such issues as some work-related task, a recent newspaper article, or the football game we will watch tomorrow. Since nothing may be the subject of our attention unless and until it triggers affect, this normative and quite friendly dissociation is the result of an intentional shift from the most logical and immanent source of affect (as would be represented by the sexual activity of the moment and the excitement of our sexual partner) to a source that produces affect of far less density and can be controlled much better. Dissociation is not intrinsically pathological, for it originates in our learned ability to make conscious decisions as to the source we will allow as a trigger for the affect of the moment.

The Blueprint

Tomkins (1962, 1987, 1991) commented that personality may be defined as the pattern of affect modulation (the differential magnification of the nine innate affects) established along the life path of each individual. He pointed out (1962) that because we have evolved to experience any of our nine affects—two positive affects that feel wonderful, one (surprise-startle) so brief that it has no flavor of its own, and six that feel simply awful—we are wired in conformance to an internal "blueprint." The human emotional blueprint guarantees that we feel best when we 1) maximize positive affect and 2) minimize negative affect; we function best when 3) we express all affect (minimize the inhibition of affect) so we can accomplish these two goals; and, finally, 4) anything that fosters these three goals makes us feel our best, whereas any force that interferes with any one or more of those goals conspires to make us feel worse. In general, psychotherapy encourages us to express affect so that we can obey rule #3, while instructing us in methods to accomplish rules #1 and #2; antidepressants fit rule #4 by removing biological impediments to rules #1 and #2. Although the effects of street drugs are transient and the postdrug experience is often far worse than the affective state for which they have been chosen as remediation, cocaine and the amphetamines increase interest-excitement directly, and heroin increases enjoyment-joy. Public entertainments provide liberal but variable doses of positive affect that also foster rule #1. Any successful form of psychotherapy must follow these four rules.

The Capacity for Intimacy

In 1991, when Tomkins was in the early stages of the lymphoma to which he succumbed midyear, Vernon C. Kelly Jr. and he worked together to take the Tomkins blueprint into the realm of interpersonal intimacy. The Kelly-Tomkins blueprint states that intimacy is a private interpersonal relationship within which two people work to 1) mutualize and maximize positive affect, and 2) mutualize and minimize negative affect. They are best able to do this when they 3) express all affect to each other (minimize the inhibition of affect) so that they can achieve the first two goals; 4) anything that interferes with any of these three rules will prevent or impede intimacy, and anything that favors the achievement of these three goals will augment intimacy. In a series of articles written for the Bulletin of the Tomkins Institute, several presentations at professional meetings, and a recent extensive chapter, Kelly (1996) has presented an entirely new system of couples therapy based on our growing understanding of affect theory and the mechanisms for empathy. After a brief anamnesis focusing on each partner’s history of affective expression, Kelly explains the affect system, teaches couples the names and functions of the nine innate affects, and helps them achieve the four goals stated in the blueprint. Emerging from this work is the understanding that the degree of interpersonal intimacy possible for any couple is dependent on the sophistication of their management of the empathic wall. The capacity for intimacy is directly proportional to the ability of individuals to accept and manage their own affects as well as the data provided by interaffectivity.

From the Tomkins blueprint for individual wellness we learn that not everybody is equally facile, equally able to deal with all of the affects. Psychoanalytic practice long took for granted that people and analysts were to a large extent fungible, that any well-trained analyst was equally competent to treat any willing patient; now we are more aware that affect-based personality factors are real, relatively immutable, and quite important in the selection of a guide. Our growing understanding of psychopharmacology proves beyond any doubt that a large fraction of those who visit us in need of help are prevented from performing the psychological tasks required for analytic work simply because neurochemical dysfunction can cause any affect to remain active independently of its normal physiological trigger, thereby dominate attention and consciousness, and force continued association to previous experiences of that affect not because it is the best way to reach understanding of the stimulus-affect-response sequences responsible for the patient’s problems but because of the more or less constant presence of that aberrant affective experience.

There are other reasons one or another affect may remain both salient and refractory to treatment. The early 1970s featured the nearly simultaneous appearance of three quite different approaches to analytic failures. Kohut (1971) asked us to focus on the development of the self system rather than the ego; Kernberg (1975) pointed out that certain unanalyzable patients seemed to cluster into the group he called "borderline" because they had failed to deal properly with the drive aggression; and Lewis (1971) asserted that it was our failure to analyze "shame conflicts" that led to case failure. I have become convinced that these three approaches hover about the locus that Lewis (1971, 1987) suggested; therapists ignorant of shame talk and write of "borderline" disorder, whereas therapists trained to recognize and work with shame often complete the treatment of such a patient without ever referencing that diagnostic label with its built-in excuses for failure. Now that most therapists are familiar with antidepressant treatment, the overwhelming majority of therapeutic failures referred to me for consultation can be traced to therapists’ ignorance of the psychology of shame (Nathanson, 1994b). The work of Wurmser (1981, 1997) demonstrated quite well that therapists who have been neither trained nor treated in the arena of shame are unable to understand, let alone treat, patients whose problems are shame-based. Morrison (1984a, 1984b, 1987) demonstrated shame as the predominant affective reaction in Kohut’s patients with problems in self development. A competent adult must be able to identify and express each of the nine innate affects in self and other, for without such ability self-development and interpersonal development must remain incomplete.

Selfobject functioning assumes and requires the ability to sense, determine the triggering source of, and provide modulation for the affect broadcast by another person. Whether the interaffectivity in question provides data for primal selfobject function so necessary to the development of an infant or the adult selfobject function that is essential for love and the nurturance of a fellow adult, the capacity for intimacy (Kelly, 1996) depends on the nature of the empathic wall to be found in whichever partner in a dyad is the receiver of the moment. Adult life as we know it would be impossible were we unable to achieve optimal and flexible immunity to the affect experienced by and therefore broadcast by others in our environment; much that goes wrong in our shared world may be traced to maldevelopment of the empathic wall in individuals and families. There is a clear line connecting Kohut’s (1971) concept of selfobject function to Basch’s (1976, 1983) work on empathy, to Nathanson’s (1986, 1992) concept of the empathic wall, to Kelly’s (1996) work on the capacity for intimacy and the repair of problems of intimacy. But now the line has grown even longer to include new work on the importance of selfobject function and the empathic wall in the larger, public group that we call "society" or "culture."

Two’s Company, Three’s a Crowd

It is easier to recognize the truth of an epigrammatic saying than to show why it works. Our moment-to-moment interactions with another person are characterized by relatively equal attention to both cognitive and affective communication. As we hear and process the other person’s statement, we monitor the affect that accompanies it. At the simplest level, we decide whether the affect is "appropriate" to the utterance. Stone (1996) reminds us that "When Meursault, the protagonist in Camus’ The Stranger [1946], tells us that ‘Mother died today. Or perhaps it was yesterday,’ it is not the content alone that disturbs us, but the way it is juxtaposed with his absence of discernible affect. Applying the conventional expectations of our culture, an examining clinician might consider this combination of affect and cognitive content to be ‘inappropriate.’ The words don’t go with the music, at least in the opinion of the observer" (p. 23). Our analysis of the match between verbal communication and expressed affect is rapid, habitual, skilled, and performed without conscious attention. The results of this analysis enter consciousness only when mismatch triggers affect that produces attention to its source; the gestalt of these skills forms a major portion of the empathic wall.

In the sort of relationship Kelly (1996) characterizes as intimate, such processes go on constantly, and the level of these skills developed by both parties is one determining factor for the limits of intimacy possible for any couple. Yet such analytic work is tiring whether in the clinical setting that is our professional life, or in the interpersonal relationship that is the center of our personal lives. What happens when we introduce a neonate into the mix? Even the happiest of couples, made up of partners who follow the four rules for intimacy, is discomfited by the arrival of a much-wanted newborn child whose unmodulated displays of all nine innate affects must take over interpersonal space previously governed by carefully established rules for the management of affective resonance. Although we take as quite reasonable that babies require the sort of attention to their affective output that makes selfobject function possible, during the period of time that they can communicate only by affective display, the mature husband-wife selfobject operations that make love so wondrous are either suspended or placed at tremendous risk. The presence of an infant forces parents to diminish the intensity of the affect they display to each other, to resonate and respond less effectively to what is displayed, and to live at least for a while in an atmosphere of vastly reduced intimacy.

Infancy actually provides a special case of the sort of interference with interaffective process produced by the introduction of any third person into a previously intimate interaction. Among adults, intelligent and capable as most of us are, the complexity of interaffective processes demanded by a three-way conversation usually results in a significant diminution of the affective intensity allowed by the participants. Empathic wall function allows us to mute the density of affect expressed, diminish the degree to which we will resonate with it, and limit our freedom to associate to information derived from affective resonance. It is the empathic wall that makes one person company and more than one companion part of a crowd, and it is the empathic wall that accounts for the way we are open in one arena and guarded in the other.

Is it then our fate to be increasingly uninvolved with others in direct proportion to the sheer number of people with whom we interact? Or, perhaps, might our reduction in affective resonance be asymptotic rather than linear so that as citizens we retain some ability to resonate with our fellows rather than withdraw to a null point at which we are incapable of feeling the world of others? Individual families are city-states, small empires that exist behind heavily defended walls within which they establish highly specific rules for the management of affect. All of us have had the experience of meeting the relative of a friend and laughing at similarities that are partly genetic trait but largely the result of the particular differential magnification of affect required by their family of origin. What we bring into the public realm is different for each of us on the basis of such early life experience. Furthermore, even though I have postulated that the development of the empathic wall is well established by three years of age (Nathanson, 1986), lessons about affect display and resonance form a significant part of kindergarten and early grade-school experience. We are educated toward a generalized, culturally acceptable standard for the empathic wall throughout early childhood.

Even though the educational establishment teaches affect modulation, it also affords instruction about the arenas within which we are allowed to express affect with more gusto. Running parallel to these lessons about decorum is a culturewide, nearly obsessional concern with athletic competition to which children are encouraged to respond with enthusiasm. Sporting events become a venue wherein the growing child is allowed to express contempt (the fusion of dissmell and anger), disgust, rage, high levels of excitement, and great joy. As we grow from school to the entertainments of the community, civic standards for the expression and resonance of affect are made increasingly clear. In a democratic society, election campaigns offer another venue within which extremes of affect may be expressed with impunity, for people can say things to each other during elections that would cause a barroom brawl or an upheaval in any other setting. The knowledge that we are allowed to express ourselves at such intensity helps convince the populace to accept a significant and perhaps otherwise unacceptable degree of restraint outside of these venues.

The Legal System

Tomkins (1991) commented that even though all innate affects arc contagious, anger resonates with peculiar ease. Angry, fighting dyads must be separated lest through contagion of affect they continue to produce more anger in each other and in those who at first were outside their quarrel. One of the most important benefits offered by the legal system is that it grinds exceedingly slowly—by delaying resolution of disputes until the protagonists have had a chance to "cool down," and by preventing the disputants from handling their own cases before the community, affect is unlikely to promote vigilante action or mob levels of dyscontrol. Eventually, cases are decided by a judge who is accepted as neutral to the process under examination, and the beliefs of each side are presented by lawyers who alone are allowed to speak with that judge. The legal system is an important part of the civic empathic wall.

Even this historically hard-won system of laws has an inherent flaw. When victims are unable to express to an empathic listener their feelings about the experience suffered at the hands of the perpetrator, they tend to feel marginalized, insignificant, unheard. Victim dissatisfaction within the public sphere is a cognate for empathic failure in the interpersonal world. Furthermore, when those accused of a crime are "protected" by their lawyers from the (often reasonable) affect of those they have injured they are thereby encouraged to remain aloof from the effects on the community of their actions. Quite naturally, defense attorneys are far more likely to show their clients how well they were served by this legal representation than to instruct them in matters of civic responsibility. Victim resentment and criminal recidivism may well be linked to this function of the public empathic wall; it is not difficult to understand both as examples of public selfobject failure.

The Family Group Conference

In 1989 New Zealand, in response to indigenous Maori people’s concerns that too many of their young people were caught up in the criminal justice or social welfare system, the government initiated a process called the "family group conference." In lieu of the formal court process, the conference brings together the extended family of an offender or a child facing removal from his or her natural home, and involves family members in the decision-making process. Victims of an offense are also invited to participate. Family group conferencing in New Zealand has resulted in a dramatic reduction of cases going to court.

In 1990, Australian police sergeant Terry O’Connell, who headed the community policing program in Wagga Wagga, New South Wales, developed a scripted version of family group conferencing. Known as the Wagga Wagga model, or in North America, the Real Justice conference, this scripted process was initially used by police officers as an alternative to referring young people to court, but has been replicated in many places around the world as a response to wrongdoing in schools, workplaces and communities. The conference facilitator invites offenders, victims, their family and friends to a meeting to see how people had been affected by the offense and how the harm might be repaired. Most notably the conference provides a forum where affect is permitted to resonate within a framework of respect and decorum. Unlike the legal system, the conference provides offenders with an opportunity to gain some empathy for those they have affected with their behavior.

Coincidentally writing in 1989, the same year that conferencing began in New Zealand, John Braithwaite, in his highly influential 1989 book Crime, Shame and Reintegration suggested that contemporary management of criminal behavior failed because it does not produce enough shame in the perpetrator, and that the proper use of shame might motivate criminals to want reconnection with the culture at large. His book was seen as an explanation for the conference process, where members of the community explain the effect of an offense on them and their lives and the perpetrator comes to understand that she or he is a part of this community and has sinned against it. Immediate to this recognition comes an outpouring of profound personal shame, after which the community gathers around and welcomes back the previously unconcerned perpetrator. Braithwaite saw here an analogue of the Japanese cultural system, which he claimed was relatively free of crime because of the degree to which shame is experienced as a toxic emotion.

In contrast to Braithwaite’s proper use of shame, hardly a day passes that one or another of the broadcast or print media does not offer a story about a judge or a jurisdiction that has begun to use shame as an intentional part of punishment. This purportedly therapeutic induction of shame has been used in many cities—those who have been convicted of driving under the influence of alcohol have been ordered to use automotive license plates bearing the world DRUNK, businesses that cheat the public have been ordered to take out highly visible advertisments in the local papers apologizing for their actions, and the names of men caught consorting with prostitutes are published in the local newspapers.

In a reintegrative family group conference, children who admit that they have committed the crime in question are shunted away from the court system and onto another track. Together with the specific victim of their action, and the families and neighbors of both victim and offender, the admitted criminal meets for a group session at which all concerned are encouraged to tell how they felt about the action under examination. Not surprisingly, the initial response of the perpetrator is often indifferent and unconcerned; even children seem unable to commit a crime when concerned about the needs and feelings of others. Yet as the conference runs on and both family groups began to speak about their estrangement from the perpetrator, that individual comes swiftly to learn that the love of the community is a deeply missed and quite important part of his or her world. With such recognition comes an avalanche of shame, after which the individual is likely to express remorse, accept the forgiveness of all concerned, and sign a document pledging to work in some way to repair or undo the damage produced by the antisocial act.

The Blueprint for Community

But why is shame so important in this process? wondered David Moore, who was teaching at the Wagga Wagga campus of Charles Sturt University when O''Connell was beginning his conferencing effort. After studying every book available on that emotion, he came to the conclusion that the shame seen in these situations was a secondary phenomenon. Primary, he came to realize, was the change in the civic empathic wall produced by the conference process. Between 1994 and 1996, Moore, O''Connell and their colleagues visited Philadelphia to discuss their system in terms of Basch’s (1976, 1983) work on affect and empathy, and that by Nathanson and Kelly previously cited. As an explanation of the Family Group Conference, I offered a Blueprint for Communities (Nathanson, 1995b), which states that a community is a public group of people linked by scripts for systems of affect modulation. It is formed and maintained by the following rules: 1) Mutualization of and group action to enhance or maximize positive affect. 2) Mutualization of and group action to diminish or minimize negative affect. 3) Communities thrive best when all affect is expressed so these first two goals may be accomplished. 4) Mechanisms that increase the power to accomplish these goals favor the maintenance of community, mechanisms that decrease the power to express and modulate affect threaten the community.

In a recent publication, Moore (1996) discusses the impact of these theories on the communities that have been served by both the Family Group Conference and now the newly developed process of workplace conferencing. It has become increasingly clear that shame is reintegrative only when it takes place in an individual who had lived outside the interaffective life of the community until returned to it through a process like the Australian system described earlier, and then experienced or recognized his or her chronic prior estrangement from empathic connection with the community as a blow that then produced shame. The mutative force is empathy, not shame, and the interaffective processes described in this communication are both ubiquitous and fractal at all levels of human interaction.


Mourning involves celebration as well as loss. The untimely death of Michael Franz Basch brings into sharp focus the number, depth, and importance of his contributions. From his early efforts to explain Kohut’s (1971) observation that the mothering caregiver is able to tune in on the world of the wordless infant by experiencing something within herself, we have developed a science of empathy that assists the repair of the most intimate human relationships and the connection to society of its most troubled and least intimate members.


Basch, M.F. (1976), The concept of affect. J. Amer. Psychoanal. Assn., 24:759-777.

———(1983), Empathic understanding. J. Amer. Psychoanal. Assn., 31:101-126.

———(1995), Doing Brief Psychotherapy. New York: Basic Books.

Braithwaite, J. (1989), Crime, Shame and Reintegration. Cambridge, UK: Cambridge University Press.

Camus, A. (1946), The Stranger. New York: Knopf.

Demos, E.V. (1983), A perspective from infant research on affect and self-esteem. In: The Development and Sustaining of Self-Esteem in Childhood, ed. J. Mack & S. Ablon. New York: International Universities Press, pp.45-78.

Freud, S. (1917), Mourning and melancholia. Standard Edition, 4:152-172. London: Hogarth Press, 1950.

———(1921), Group psychology and the analysis of the ego. Standard Edition, 18:67-143. London: Hogarth Press, 1955.

Kelly, V. C. (1996), Affect and the redefinition of intimacy. In: Knowing Feeling: Affect, Script, and Psychotherapy, ed. D.L. Nathanson. New York: Norton, pp.55-104.

Kernberg, O. (1975), Borderline Conditions and Pathological Narcissism. New York: Aronson.

Kohut, H. (1971), The Analysis of the Self. New York: International Universities Press.

Lewis, H. B. (1971), Shame and Guilt in Neurosis. New York: International Universities Press.

———(1987), Shame and the narcissistic personality. In: The Many Faces of Shame, ed. D.L. Nathanson. New York: Guilford, pp.93-132.

Moore, D.B. (1996), Illegal action—official reaction. In: Knowing Feeling: Affect, Script, and Psychotherapy, ed. D.L. Nathanson. New York: Norton, pp.346-378.

Morrison, A.P. (1984a), Shame and the psychology of the self. In: Kohut’s Legacy, ed. P.E. Stepansky & A. Goldberg. Hillsdale, NJ: The Analytic Press, pp.70-91.

———(1984b), Working with shame in psychoanalytic treatment. J. Amer. Psychoanal. Assn., 32:479-505.

———(1987), The eye turned inward: Shame and the self. In: The Many Faces of Shame, ed. D.L. Nathanson. New York: Guilford, pp.271-291.

Mowrer, D.H. (1939), Stimulus response theory of anxiety. Psycholog. Rev., 46:553-565.

Nathanson, D.L. (1986), The empathic wall and the ecology of affect. The Psychoanalytic Study of the Child, 41:171-87. New Haven, CT: Yale University Press.

———(1987), A timetable for shame. In: The Many Faces of Shame, ed. D. L. Nathanson. New York: Guilford, pp.1-62.

———(1989), Denial, projection, and the empathic wall. In: Denial, ed. E. Edelstein, D.L. Nathanson & A.M. Stone. New York: Plenum, pp.37-55.

———(1992), Shame and Pride: Affect, Sex, and the Birth of the Self. New York: Norton.

———(1994a), The case against depression. In: Knowing Feeling: Affect, Script, and Psychotherapy, ed. D.L. Nathanson. New York: Norton, 1996, pp.388-394.

———(1994b), Shame, compassion, and the "borderline" personality. Psychiat. Clinics North Amer., 17:785-810.

———(1995a), Pain and the affect system: Psyche, script, and soma. Bull. Tomkins Institute, 2:13-15.

———(1995b), Crime and nourishment: Sometimes the tried and true becomes the tired and false. Bull. Tomkins Institute, 2:25-30.

Nathanson, D.L. & Pfrommer, J.M. (1996), Affect theory and psychopharmacology. In: Knowing Feeling, ed. D.L. Nathanson. New York: Norton, pp.177-190.

Scheler, M. (1912), The Nature of Sympathy. Hamden, CT: Archon Books, 1972.

Spitz, R. (1965), The First Year of Life. New York: International Universities Press.

Stone, A.M. (1996), Clinical assessment of affect. In: Knowing Feeling, ed. D.L. Nathanson. New York: Norton, pp.22-36.

Sullivan, H.S. (1954), The Psychiatric Interview. New York: Norton.

Tomkins, S.S. (1962), Affect Imagery Consciousness, Vol. I. New York: Springer.

———(1963), Affect Imagery Consciousness, Vol. II. New York: Springer.

———(1981), The quest for primary motives: Biography and autobiography of an idea. Personality Social Psychol., 41:306-329.

———(1987), Shame. In: The Many Faces of Shame, ed. D.L. Nathanson. New York: Norton, pp.133-161.

———(1991), Affect Imagery Consciousness. Vol. III. New York: Springer.

Wurmser, L. (1981), The Mask of Shame. Baltimore: Johns Hopkins University Press.

———(1987), Shame: The veiled companion of narcissism. In: The Many Faces of Shame, ed. D.L. Nathanson. New York: Guilford, pp.64-92.


Restorative Works Cover
Restorative Works Year in Review 2023 (PDF)

All our donors are acknowledged annually in Restorative Works.