D M Black

Sample psychoanalytic paper (1)

Here is a sample essay on a psychoanalytic theme, first published in 2004 in the International Journal of Psycho-Analysis, 85 pp 579-595

Sympathy reconfigured: Some reflections on sympathy, empathy and the discovery of values

David M. Black

Abstract. Two words, sympathy and empathy, are commonly used to describe three distinguishable things. These are: i) an elementary, involuntary capacity which puts us in touch with the emotional state of another; ii) the use of ‘trial identification’ to discover, consciously or unconsciously, the emotional state of another; iii) the affect of compassion. Because these three usages have not been clearly sorted out, and because the word sympathy has been disparaged, empathy has been overused, and a variety of technical terms (including inter subjectivity, recurrent primary identification, projective identification, alpha function etc.), all of which have important specialised applications, have been used confusingly to describe functions at a much higher level of generality. This paper attempts to sort out the three meanings with reference to the history of the two words, and also to show that the use of clear general terms gives us a more intelligible linguistic base from which other matters can be considered: in particular, whether an origin can be discerned for judgements of value other than that of superego internalisation.


Because psychoanalytic theory started out from Freud's early one-body neurologically based picture, and then, in Freud's later work and the work of subsequent analysts, tried to reach out from there to give an account of the irreducibly social nature of human psychology, we can find ourselves confused, sometimes, about the order and logical hierarchy of our concepts. We catch ourselves trying to explain simple notions by more complex ones. In this paper, I am going to suggest that some of the explanations we give of social interaction—in terms of such concepts as intersubjectivity, alpha-function, projective identification and mentalisation—have lacked, or, to be more accurate, have failed to recognise, an elementary building-block. The ordinary name in English for this building-block is sympathy; but to use this word gives rise to certain problems.

I shall start with some reflections about the current state of the word

My remarks will relate to the world of psychoanalysis, but I think they also apply more generally, and I shall suggest that, if so, they indicate something more important than just a curious byway in the history of the English language.

sympathy is a word psychoanalysts avoid. If an analyst uses it inadvertently, he often ‘corrects’ it and says ‘empathy’ instead. Empathy, by contrast, is a word we use with pride. We feel that by empathy we make a trial identification with the other, without losing our secure stance in ourselves, and as a result of empathic contact we can relate our interpretations accurately to the patient's internal state. Kelman (1987) spoke of ‘the current universal approbation accorded to empathy’.

The word
sympathy has also suffered from a belief that it implies a wish to help, and we are trained to be suspicious of such a wish. Perhaps this suspicion has created a tough- mindedness, in some parts of our profession, which has made it hard to use the word sympathy without shame at its apparent softness and lack of direction. Freud (1912) explicitly recommended psychoanalysts ‘to model themselves ... on the surgeon, who puts aside all his feelings, even his human sympathy’.

So fundamental a notion, however, has inevitably been a silent presence in much psychoanalytic thought, particularly about babyhood. Bion's account of maternal reverie (
1962, pp. 35-6) clearly implies the existence of sympathy, but the word reverie, which appears to describe something taking place entirely within the mother, doesn't help us to understand the complex communication which reverie can arrive at (see my first case example below). The many references to a ‘genetically early state’ of ‘primary confusion of self and object’ (Sandler, 1988) are obscure because they leave unexplained the mechanisms by which such a confusion could occur. Stern's (1985) important notion of ‘affect attunement’ presupposes a sympathetic connection between mother and child.

One of the clearest accounts in the analytic literature of what I shall call the primary meaning of sympathy appears in a discussion by Sandler
(1993). Sandler describes walking in a crowded street, and seeing someone walking ahead of him stumble. He instantly corrected an impulse in himself to stumble. He speaks of this as an example of ‘recurrent primary identification’ or ‘unconscious temporary mirroring’, and suggests it is pervasive in our communications. He claims that much of what is called ‘projective identification’ is, in fact, attributable to this other sort of identification, and that what is properly called projective identification (by the patient) will also require the presence of this primary identification (by the analyst). Interestingly, he quotes psychologist William McDougall, who in 1909 described the same phenomenon and called it ‘primitive passive sympathy’ (cited in Sandler, 1993, p. 1101). But, for his own account, Sandler prefers the more technical phrase.

When we find the word
sympathy used without embarrassment, it is often in its adjectival or adverbial forms. Developmental psychologists Trevarthan and Aitken, for example, in defining ‘primary intersubjectivity’, describe experiments which ‘show that the young infant has expectations of the emotional quality of the engagement and the normal contingencies of a sympathetic adult response ...’ (2001, p. 9). They also speak of it in the negative: infants are traumatised if they meet ‘unsympathetic and inappropriately timed maternal behaviours’ (p. 9, my italics).

History of sympathy

Let me now look back into the history of the word sympathy. It derives from the Greek, the sym- being with, and the pathy coming from pathos, which implies suffering. The word pathos also stands behind our familiar words patient, passivity, passion etc. I shall spend some time on this history, because I believe our present usage of the word sympathy has cost us something very important. The word suffering, interestingly, has followed a similar path. Its primary meaning is something more like experiencing: what I do not do actively, I suffer, I experience, passively (receptively). When Goethe undertook to describe the origin of the different colours, he spoke of ‘the deeds and sufferings of light’ (Bortoft, 1996, p. 46). He didn't mean that light can suffer pain, as we might expect with that word now; he meant merely to describe the way light can be altered by something other than itself, for example by the thickness of the atmosphere.

sympathy in its primary meaning is not to do with sharing pain, but with sharing the passivities of human experience. Its importance was briefly recognized by some of the major thinkers of the 18th-century Scottish Enlightenment, but disappeared soon after because, I suggest, of the difficulty of holding on to its primary and fundamental meaning. It was described very clearly, however, in the early writings by David Hume. In his major work A treatise of human nature, he writes,

“We may begin with considering anew the nature and force of
sympathy. The minds of all men are similar in their feelings and operations; nor can any one be actuated by any affection, of which all others are not, in some degree, susceptible. As in strings equally wound up, the motion of one communicates itself to the rest; so all the affections readily pass from one person to another, and beget correspondent movements in every human creature” (2000, p. 368).

You will notice that this definition makes no reference to pain, and Hume certainly does not intend his definition to relate to any one subdivision of human experience. I shall take the liberty now of quoting him at length, because I think Hume here is defining something fundamental. He writes:

“No quality of human nature is more remarkable, both in itself and in its consequences, than that propensity we have to sympathise with others, and to receive by communication their inclinations and sentiments, however different from, or even contrary to our own. This is not only conspicuous in children, who implicitly embrace every opinion propos'd to them; but also in men of the greatest judgment and understanding, who find it very difficult to follow their own reason or inclination, in opposition to that of their friends and daily companions. To this principle we ought to ascribe the great uniformity we may observe in the humours and turn of thinking of those of the same nation; and ‘tis much more probable, that this resemblance arises from sympathy, than from any influence of the soil or climate, which, tho’ they continue invariably the same, are not able to preserve the character of a nation the same for a century together. A good-natur'd man finds himself in an instant of the same humour with his company; and even the proudest and most surly take a tincture from their countrymen and acquaintance. A chearful countenance infuses a sensible complacency and serenity into my mind; as an angry or sorrowful one throws a sudden damp upon me. Hatred, resentment, esteem, love, courage, mirth and melancholy; all these passions I feel more from communication than from my own natural temper and disposition. So remarkable a phaenomenon merits our attention, and must be trac'd up to its first principles” (p. 206).

The point I wish to emphasise from Hume is that sympathy, in his sense, is not a feeling. It is a ‘propensity’, a spontaneous capacity, to pick up the feelings of others. Hume has been criticised by modern philosophers, for example Rawls, who says this is not an account of sympathy ‘as we normally understand it’, but rather an account of ‘imparted feeling’ (
2000, p. 86). But this I think is a mistaken projection backward of our modern linguistic usage. Hume is using the word sympathy more, not less, accurately than our modern idiom, and our loss of his precise meaning leaves us floundering in a welter of technical terms (intersubjectivity, recurrent primary identification, mirroring, resonant cognition, alpha-function, projective identification, mentalisation), all of which have important specialised applications but none of which is fit to fulfil the elementary and general role that Hume's sympathy occupies.

This account of sympathy as a capacity, quite distinct from the feeling that is often given the same name, is one that proved hard to hold on to. In Hume's later writings he seemed to lose hold of it himself, preferring the even more ambiguous term ‘fellow feeling’, and his follower Adam Smith
([1759] 1976), the founder of the science of economics, although he famously based his whole understanding of what he called the ‘moral sentiments’ on sympathy, meant by sympathy something markedly different from the capacity. Smith's concern was with ‘mutual sympathy’, the basis in his view of honest friendship and social approval; we strive to achieve such sympathy with others, he says, and where it cannot be reached the painful emotions of disapproval are activated (Phillipson, 1983). Because Smith's account of sympathy has been widely quoted, it has tended to push Hume's earlier version into the background.
Interestingly enough, more than 100 years after Hume wrote his
Treatise, when Darwin came to think about the development of the distinctive human characteristics, he too put what he called ‘the all-important emotion of sympathy’ (1871, 106) squarely in the centre of the picture. He saw sympathy as widely present in the higher animals (he mentions dogs, monkeys, elephants etc.) and also in certain birds, including pelicans and crows. Darwin was a very good amateur psychologist, and many of his formulations remain very suggestive. For example, he writes,

“Even when we are quite alone, how often do we think with pleasure or pain of what others think of us—of their imagined approbation or disapprobation; and this all follows from
sympathy, a fundamental element of the social instincts. A man who possessed no trace of such instincts would be an unnatural monster ...“ (112, my italics).

And again:

“Many a civilised man, or even boy, who never before risked his life for another, but full of courage and
sympathy, has disregarded the instinct of self-preservation, and plunged at once into a torrent to save a drowning man, tho' a stranger ... Such actions ... appear to be the simple result of the greater strength of the social or maternal instincts than that of any other instinct or motive ...” (110, my italics).

In these passages, Darwin interestingly links the motive of sympathy with the ‘maternal instincts’, and, although he doesn't spell out how boys or men might act from maternal instincts, we can glimpse behind his writing some shadowy notions of internalisation and transference. In the first quotation, we can see the importance of sympathy in the formation of internal objects, a notion not available to Darwin but clearly anticipated by him here.

Darwin, however, was not a philosopher and his use of language lacks Hume's precision. Much of the time his word
sympathy has Hume's meaning, and he quotes Hume at one point (109n), but often, as in the passages quoted, it gets mixed up with the more modern meaning, of an emotion of warm concern, and he doesn't distinguish between these two usages.

Sympathy and empathy

The recognition of sympathy in animals, or pre-verbal infants, is of particular value to my argument because it allows the differentiation of sympathy from empathy to be made rather clearly. Empathy is a modern word, coined as a translation of the German Einfühlung by American psychologist Edward Bremner Titchener in 1909. George Pigman (1995) has written an excellent summary of the history of Einfühlung, which I draw on in my next few paragraphs.

In 1909,
Einfühlung was a relatively new term in German, used first in 1873 by art- historian Robert Vischer. It arose out of a long history of discussion by German thinkers of the way in which we can be moved by aesthetic perceptions. In 1774, for example, Goethe's friend Herder had urged those wishing to understand the world of the Ancients to ‘fühle dich in alles hinein’, feel yourself into everything.

Theodor Lipps (1851-1914), a philosopher admired by Freud (and quoted wrily by Freud at one point to Fliess as having anticipated many of Freud's own discoveries), gave a central place to Vischer's concept. Writing in 1903, Lipps spoke of ‘an arrangement of my nature which is not further describable’ which causes me both to have an impulse to mimic the facial expression of someone in front of me, and to tend to experience the affective state represented by that expression. This ‘arrangement’, though described in more detail, is very close to Hume's account of sympathy as a spontaneously acting capacity. According to Lipps, when I ‘feel myself into’ the resulting psychic experience, I achieve
Einfühlung. Lipps believed Einfühlung should be extended from its initial aesthetic application and become a fundamental concept in psychology and sociology—that Einfühlung ‘is what enables us to realise that we have selves and that other selves exist’ (Pigman, 1995, 242).

Freud is often thought to have given little place to
empathy. This, however, is a misapprehension. As with Nachträglichkeit, Einfühlung and its related forms were used on quite a number of occasions by Freud—some 20 in all—but were not consistently translated by Strachey in the Standard Edition. (It seems likely that Strachey objected to the words empathy and empathise on aesthetic grounds.) One occasion when Freud used Einfühlung is in a famous passage in his paper ‘On beginning the treatment’: ‘It is certainly possible to forfeit this first success if from the start one takes up any standpoint other than that of sympathetic understanding, such as a moralising one’ (1913, 140).

‘Sympathetic understanding’ is Strachey's rendering here of
Einfühlung, doubly misleading, as Pigman points out, because it fails to reveal Freud's positive valuation of empathy, and also fails to distinguish it clearly from sympathy in the sense of ‘sympathising’. (Pigman's article goes on to discuss several of Freud's uses of the concept, emphasising in particular that for Freud, unlike Lipps, empathy had a strongly intellectual flavour, and relates to the process of putting oneself, consciously or unconsciously, into someone else's position in order to understand him. This brings Freud closer than Lipps to the prevailing modern usage of the word empathy.)

Since Freud, many psychoanalysts have used the term, with greater or less concern for these shades of meaning. Kohut, in particular, gives a central place to
empathy, defining it as ‘a mode of cognition which is specifically attuned to the perception of complex psychological configurations’ (1971, 300). To call it a mode of cognition is suggestive, and relevant to the important issue of accuracy in sympathetic or empathetic perception, but Kohut doesn't make the kind of distinctions among these notions that I want to identify.

Hinshelwood sets out to be more precise. In
A dictionary ofKleinian thought, he defines empathy as a form of projective identification, ‘one of those benign forms of projective identification which can be included in “normal projective identification”’(1989, 295).

Projective identification was originally defined by Klein
(1946) as a phantasy. It was a phantasy used as a defence to cope with certain parts of the self which were felt to be unbearable and therefore projected out and imagined to be located in the other, with consequent weakening and depletion of the self. This notion was subsequently vastly expanded by Bion and others, who suggested projective identification was also a part of normal communication, and particularly so in the case of pre-verbal infants and their mothers. Bion (1962, 31-7) speaks of the trauma a baby suffers with a mother who is unable to receive projective identifications.

The notion of projective identification has been developed in great and sometimes controversial detail, especially but not exclusively in the Kleinian literature (see
Sandler, 1988). Hinshelwood's account of empathy states that it is a certain form, namely a benign form, of projective identification, in which one person projects his capacity for self-perception into someone else's situation in order to ‘gain’, as he puts it, the other person's experience. He emphasises that in empathy there is ‘no loss of reality, no confusion of identity’ (1989, 296).

The phrase ‘no confusion of identity’ implies that this is a sophisticated mental operation in which the two persons involved are very clearly separate but one is consciously setting out to conceive the other's mental landscape. We might think of psychoanalyst and patient, or salesman and customer, and there is something analogous in chimpanzees, for example. A small chimpanzee who has spied a banana, in the presence of a larger chimpanzee who doesn't know it is there, will skilfully position himself between the large chimpanzee and the banana so that the large chimpanzee can't see it; and he will nonchalantly look around and pretend to be preoccupied with other matters until the large chimpanzee quits the scene. Absent the competition, the small chimpanzee at once grabs the banana. The mental operations involved in such a deception are quite elaborate; they include a capacity to imagine himself in the other chimpanzee's place, in a physical sense, and also to be aware, in some sense, of the other chimpanzee's desires and thinking processes, while postponing gratification and retaining a very clear contact with his own self-interest.

What I am calling
sympathy is different from this. Self and other are not clearly distinct, and I am inclined to question whether what sympathy describes is correctly thought of as a mental operation only. It may be an operation that engages the soma as well as the psyche of both parties, perhaps at the level of ‘mirror neurons’ in the frontal cortex (Damasio, 2003), which work by spontaneously initiating mimicry of a perceived affective display. (It is interesting to compare this modern neuroscientific account with Lipps's introspective understanding of Einfühlung in 1903, above.) Sympathy, as I understand it, makes empathy possible, but they operate at different levels of sophistication; empathy is a sophisticated and conscious act, which uses as a tool the elementary and inescapable (involuntary) capacity that I, following Hume, would call sympathy. Because sympathy is with the other as subject, rather than as object, it relates in an inchoate way to the other as ‘whole-object’ rather than part-object (but this language is, of course, odd when used in relation to subjects).

For clarity's sake, we need to make a clear distinction between the two usages of the word
sympathy. I shall call Hume's usage, the capacity, sympathy sense one, and the modern usage, the emotion, sympathy sense two. To dramatise and insist on the distinction, we might say that the professional torturer, tormenting his victim, must be very sympathetic to his victim. That is, in sense one. In sense two, of course, he is horribly unsympathetic. He knows what it feels like, but he knows without compassion.

It would appear also to be true that the torturer has a certain empathy with his victim. But that sounds too paradoxical, and I think we would not say it. Perhaps empathy entails not only a knowing of the other person's feeling, but also an identification with the other as the one having that feeling, and therefore it implies that sympathy in the second sense, human concern or fellow feeling, has not been extinguished.

Let me summarise and draw out a few points relating to these reflections so far.
First, the word
sympathy has become disparaged, although it still gets smuggled in usefully in its adjectival forms. If we look at it historically, however, we find it has two distinct and important meanings: 1. A spontaneous capacity to be directly affected by the feeling state of others. It is this capacity, well described by Hume in his early writings, which enables us, for example, to recognise instantly, and very often correctly, what Stern (1985) has called ‘vitality affects’, the unverbalised emotional messages carried by every detail of another person's posture, gestures, hesitations, use of eye-contact, tones of voice etc. Neuroscientist Damasio (2000) has described the same phenomenon using the term ‘background emotions’. Sympathy in this sense is already present in babies and, to some extent, in animals. It is a capacity, not an emotion. 2. A warm concern for the feelings of others. Sympathy in this sense, also called compassion, is an emotion, or a range of emotions, akin to sorrow and belonging with the depressive position group, and like other emotions it can be highly developed, repressed, split off etc.

empathy—although the word is often used where sympathy would be more accurate—is a sophisticated act of the imagination, a ‘trial identification’ done by someone who is consciously relating to another's mental state.
Third, empathy and sympathy sense two both depend on the elementary capacity, sympathy sense one. Like our other perceptual capacities, sympathy (sense one) is often remarkably accurate, although it can be mistaken, misled etc. It is educated and developed by emotional experience, and especially by experiences of ‘affect attunement’
(Stern, 1985) with caring others in infancy.
Fourth, the relation between sympathy (sense one), the capacity, and sympathy (sense two), the emotion, is complex, and raises questions of psychological health, normality etc. We might caricature two extremes by saying that the charming sociopath is strong in sense one, but devoid of sense two, whereas the woman who commits altruistic surrender
(A. Freud, 1936) is so overwhelmed by sense two that she neglects her own proper needs.

There is probably no intrinsic reason why the conceptual territory should be divided up between these words in the way I am recommending. The initial use of the word
Einfühlung, by Vischer and Lipps, is as close to my sympathy sense one as it is to empathy. But, given the use of sympathy by Hume and Darwin, and the later history of Einfühlung/empathy in the hands of Freud and later psychoanalysts, I think my distribution of the words is intelligible historically, and does as little violence as possible to our ordinary verbal expectations.

I have said that I thought the fate of the words
sympathy and suffering may have a wider significance. Both words, which initially implied simply ‘experiencing’, have come to mean something more akin to ‘experiencing pain’, ‘suffering’ in its modern sense. It is as if passivity/receptivity, the necessary condition of receiving experience, has been given an increasingly negative reading; as if activity alone can give happiness. Here, of course, we make contact with a familiar theme of classical psychoanalysis, summed up by Freud (1937) at the end of his life as the ‘repudiation of femininity’ in both sexes. It may well also have a sociological significance, to do with fear in an increasingly individualistic and competitive society of motives that do not contribute to self-assertion.

First case example

Mrs A was a woman in her forties who came to see me because of depression. She was a tall, strongly built woman, an opera-singer by training, who grew up with a violent policeman father and a sensitive religious mother. A much-wanted son was born soon after her. Promiscuous and lonely during her teens, she later became depressed, abandoned her budding career and married an alcoholic writer. Her initial attitude to me was very respectful and deferential. After about seven years, with her life situation greatly improved and the prospect of ending analysis beginning to be spoken of, she became enormously angry with me. In session after session, she would be speechless with rage—sometimes lying in silence for the entire 50 minutes—and in those moods she would experience any remark by me as intolerable provocation. She would shout at me that she hated me, she saw me as smug, and she hated me in particular for saying that she didn't want to face reality. What made me think I knew what reality was? Her reality was not the same as mine.

I interpreted that she idealised the position of the baby. I said she thought that the truly desirable thing was to be a baby, and the thought of growing up, and developing, and of eventually leaving analysis, seemed to her nothing but loss; and, if I thought it was desirable, and that it was a good thing that we were working towards it, then I was stupid and arrogant and entirely out of touch with reality. I didn't make all these interpretations at once, but this was the tenor of our exchanges in many sessions. Typically, when the silence broke, and she raged and stormed at me for most of a session, I would touch some chord in her, or she would touch some chord in herself, and she would break down in tears and self-reproaches. We might then have two or three sessions in a more ordinary frame of mind, and then the rageful silences would recommence.

As the months went by, I found this pattern increasingly hard to bear. She was coming five times a week, never missed a session and had a powerful presence. I worried that nothing I had said seemed helpful. I began to feel I had exhausted what I could understand of the reasons for the pattern, and I found it hard to be interested in repeating my interpretations. I felt punished unjustly by the enormous silences, and then often shocked and disconcerted by the rage and shouting, which sometimes frightened me, and sometimes reduced me to silence. One session, after she had again bellowed her hatred of me, we fell into what felt like an angry, silent, mutual stand-off. I felt that anything I might say, whatever its content, would be heard, and very possibly would be said, as an angry, patronising retaliation.

In the 20 minutes or so that followed, I found the tension in me ebbing. My thoughts wandered. I suddenly came to with a shock and tried to reattend to my patient, who was lying there silently. Then I remembered that wandering thoughts may be of interest, and wondered what it was I had been thinking about.

I realised I had been thinking of an episode from my adolescence, which I don't believe I had consciously remembered for decades. I had been hitchhiking alone through France in the late 1950s, when British teenagers were often treated with impressive contempt by the local people. One hot day, feeling angry and alienated, I had decided to take a train to a remote country area, and got off at random at a tiny station. Two or three locals got off as well, and preceded me down the dirt path to the little gate that was the exit from the station.

A girl about my own age, 17 or so, had come out to the gate to collect the tickets. I presume she was the stationmaster's daughter; probably this was the one train of the day. In my ill temper, and with the many pockets of the back-packer, I couldn't find my ticket, and angrily flung the things I could find—keys, cash—on to the ground. Without hesitation, she kneeled down, picked them up and politely gave them back to me. I found my ticket and handed it to her.

When I thought about this episode, now, in the session, I had an enormous sense of gratitude to this girl. She could easily have treated me with contempt or mockery, and instead she had treated me with an entirely unexpected quality, which I experienced as a straightforward, friendly acceptance of my feeling without even needing to comment on it. I had been in an infantile mood, and she had shielded me from the possibly humiliating consequences.

At this point, in the session, I suddenly saw what was happening with my patient in a quite new light. I became capable, not just of making interpretations to her, but also of being more ‘affect-attuned’ to her. I was able to say with genuine feeling how awful it was to feel raging and excluded, and at the same time to know that the feelings were unjustified, and yet be helpless to change them. I became able, so to speak, to pick up, and in the course of the ensuing work to hand back to her, important contents that belonged to her and which she needed. No single interpretation, of course, can alter a pattern that is so deeply imprinted, but from then on, remembering my French experience, I more often felt able to speak to her in a somewhat different manner. Over the next few months, the pattern gradually loosened, and termination, when it finally came, was manageable and worked towards.

If we enquire into what was happening in this situation, it seems to me that notions of empathy or projective identification account for only parts of it. Empathising is what I had been trying to do previously: I had been trying to make a ‘trial identification’ with my patient, and out of that I had formulated my various interpretations to do with her longing for the position of the baby and the bleakness she felt at developments which might be thought of by some, perhaps by me, as progress. I don't think these interpretations were wrong, but they were counterproductive because their somewhat abstract style conveyed the ‘adult’ attitudes that were being commented on; they were made by someone who was apparently unbothered by the conflict they were describing, and as a result they left the patient feeling exposed and infantile, rather than held and understood.

Projective identification takes us further, in allowing us to understand a part of the interaction. The patient's rage, which reduces me to silence, can be understood as a projective technique, intended to evacuate an intolerable affect, a sense of inferiority and helplessness, and, when I finally can find nowhere further to go with interpretations, the projection has taken place. The attempt at evacuation has no doubt failed, but in the process a communication, the imparting of a similar feeling to the other, has been achieved. This by itself, however, only leads to stalemate. What then occurs, the wandering thoughts or reverie
(Bion, 1962), which arrives at a specific autobiographical narrative (belonging to the analyst, not the patient), goes beyond being a projective process by the patient. It reconfigures or recontexts the newly acquired content, the imparted affect, so that the receiving mind becomes able to view it in a different light, or, in this case, to get access to an object—the girl in my story—who embodies some of the containing capacities the situation calls for.

I am suggesting that
sympathy sense one, the capacity, is what guides this reverie, which is not exactly unconscious but takes place in a state of distraction. Whether one should say this sympathy is mediated by somatic signals, or by intuition in the sense of subliminal perceptions, I am uncertain. Very probably the answer will turn out to be both.

The attempt at
empathy, which views the patient's mental state as an object for understanding, can sometimes be used as a defence against sympathy, which involves the acceptance of the patient as a subject like ego. If that were happening in this case, it would help to explain why the process of working through became so stuck, and why impasse and reverie were necessary before progress could be resumed.

Two claims for the importance of sympathy

I am going now to make two large claims for sympathy in sense one. First, that it is an ego-function (though it can be unconscious) and makes an important contribution to our perception of reality. For example, it improves the accuracy of our social perceptions. We can imagine someone, an overworked nurse perhaps, who might wish to help but who might also be unsympathetic (sense two). Such a person might indeed help, but less accurately than a comparable nurse who has the time and emotional openness to use her capacity for sympathy. There is also a parallel in certain scientists, such as Einstein or plant geneticist Barbara McClintock, who made impressive discoveries which were only subsequently supported by adequate evidence, and who spontaneously described the quality of attention they brought to their subject-matter as sympathetic. When Einstein described the process by which he reached his ideas, he said, ‘To these elementary laws there leads no logical path, but only intuition, supported by being sympathetically in touch with experience’ (quoted in Keller, 1983, 145, my italics). The power of sympathy in such immensely gifted individuals can cause them to seem almost clairvoyant. However, the idea of sympathising with something inanimate, such as a beam of light or the genetic material of maize, may be a metaphor; it awaits further consideration. (It is interesting to recall that the word Einfühlung originated in aesthetic philosophers such as Vischer and Lipps.)

If sympathy sense one, the capacity, can enhance accuracy, sympathy sense two, the emotion, may impair it. Freud gives correct guidelines, I think, with respect to both. His ‘evenly hovering attention’ creates the conditions for sympathy sense one; his criticism of therapeutic zeal enjoins wariness of sympathy sense two.

The second large claim I want to make for sympathy sense one is that it is the ultimate basis for what is objective in our perception of moral value. This argument needs to be made at greater length than I have space for in this paper. But essentially I would again follow Hume's argument in the
Treatise, which is carefully constructed and properly psychological. By itself, Hume says, sympathy is prejudiced in favour of those near us, or those resembling us; values depend on an extension of sympathy by imaginative acts in which we put ourselves in fantasy close to events that in reality are far away. For example, if I am in London, and I hear of suffering caused by floods in Pakistan, I may not immediately have a reaction of concern; it is only when I imagine that suffering (that is to say, imagine it as if it were nearby), and then sympathise with myself and the reaction I would feel if the floods and the suffering people were directly present to me—only then do I have what I feel to be an appropriate reaction. It is important that in Hume's account the sympathy is not in the first instance with the sufferers (that would be sympathy sense two); it is with myself in an imagined position in which I can be directly affected by the events. He later goes on to derive from this projected image of ‘myself’ an impartial person, a ‘judicious spectator’ who can be imagined as present at all events, including those involving myself directly, who would have appropriate and proportionate responses free from the inevitable biases of any actual person. Not just the fact, but also the appeal, of justice can be understood from our sympathy with this judicious spectator.

This account of moral values has several impressive advantages. It is compatible with psychoanalytic thinking in that it grounds values firmly in emotional experience, rather than in an intellectual operation such as a utilitarian calculus, or in a religious revelation. But it improves on many psychoanalytic accounts by making possible a clear distinction between morality derived from the superego, that is, from internalised precepts and prohibitions, and morality based on conscience, that is, on one's own capacity for sympathy, imagination and reflection. It makes comprehensible the reality that our best values do not always determine our actions, and also that we are aware of an emotional response of guilt, ruefulness or remorse, when this is the case. If, as Britton
(2003) has suggested, ego needs to reclaim the function of judgement from superego, this account indicates an intelligible basis on which it might do so.

It also gives us a way of understanding the feeling we have that certain developments over time, in which the ordinary moral perceptions of society are criticised and eventually modified, may represent ‘progress’ or be a backward step. For example, when the slave trade became morally repugnant to the influential classes of Europe, or when the human rights of women, children or racial or sexual minorities are asserted, we are apt to feel (though not everyone feels) that something ‘progressive’ has occurred. Hume's theory gives us a way to understand that this feeling can have an objective component and may not be merely dependent on subjective preference.

It is not possible, I think, to found the objective component of values on a mental mechanism, such as projective identification or recurrent primary identification, because what is required in addition is a perception, in some way, of reality.
Sympathy in what I have called sense one, the capacity, enables us to make an accurate appraisal of a certain sort of reality, namely another person's psychic reality and subjective experience. Freud (1900) in a bold and very interesting image described consciousness as a ‘sense organ for the perception of psychic qualities’. I am tempted to use the same imagery and speak of sympathy sense one as a ‘sense organ’ for the perception of the psychic states of others. (To arrive at proper values, of course, requires a great deal more than merely sympathy sense one. Imagination, dialogue and conscious reflection are also required.)

In successful psychoanalytic work, developments also occur in the patient's sense of values. We often speak of such changes in terms of an enlarged capacity for intimacy, more genuine toleration of difference, an increased capacity to process emotional experience before reacting to it, and so on; or we use powerful if somewhat obscure metaphors such as a capacity for greater depth in relationships or understanding. Tolerance is a particularly important notion in our hopes for the gains from analysis. Klein writes, ‘The criterion of all later capacity for adaptation to reality is the degree in which (the child is) able to tolerate the deprivations that result from the Oedipus situation’ (
1926, 129). These painful deprivations include separation, feelings of exclusion, the knowledge that one is not always loved, or always lovable, and that loved people have loving relations with people other than oneself. Nowadays, we are likely to further conceptualise all these changes in terms of an internalisation of more tolerant objects that will include aspects of the analyst's functioning during sessions.

Changes of this kind can also be seen as extensions of our capacity for sympathy, on which can be built a recognition of the reality of other people and their lives, and of ourselves as not only the peremptory all-dominating ‘I’ of narcissism but also as ‘one among many’. Put like that, we are able to see that the general direction of the developments hoped for from analysis is also the general direction of the sort of moral developments we call progressive in the public sphere. Almost any reasonably successful piece of analytic work would illustrate this; I shall attempt to do so with the following clinical vignette.

Second clinical example

Ms B came for help initially in her mid-fifties, and continued to come until she was in her early seventies. Her background was traumatic. She believed she had been sexually abused by her father as a small child, and in latency she was exposed to her parents' intercourse in grossly inappropriate ways, lying awake as they had forceful sex in the same room and on one occasion in her bed, with her in it. Aged 12, she witnessed her father's violence and sarcasm towards orphaned children whom the family took in for a charitable organisation; the father refused to allow one boy, who was sick, to go to hospital, and the boy died.

She grew up isolated and depressed. In adolescence, her relations with boyfriends were destroyed by her compulsive jealousy, which would arise without evidence and quickly reach delusional intensity. In her early twenties she became depressed and was hospitalised for a year. She trained as a librarian and managed to hold down a job in a small provincial town. In her forties she was again in psychiatric hospital with depression, following a long string of failed relationships with men.

When she came to me in her fifties, she professed extreme contempt for the ordinary values of society. No word in the language, she told me, was so ridiculous as the word ‘wife’. The best position for a woman was to be unmarried and hunting outside the camp. She had no wish for a loving relationship with a man, merely for transient sexual encounters. Such things were said with a mixture of anger, a certain grandiloquence and a kind of sad puzzlement, as if she didn't quite convince herself and yet was describing what was actually her experience. She spoke of suicide sometimes as an attractive option. She evoked a powerful feeling in me of concern for the extreme isolation of her life.

Despite her conscious memories of her past, her transference was curiously trusting. Even when, early on, she found it hard to approach the consulting room because she was besieged with images of me as a sexual abuser, she nevertheless told me of these thoughts very directly, and was calmed and relieved when she had done so, as if from somewhere she retained a memory of a good object that could be trusted.

Over the 16 years of the therapy (she never came more than twice a week), she began to write autobiographical short stories, and in her late sixties she finally published a collection of them. Publication was hugely significant to her. She felt she had lived her life skulking in hidden corners, and now she had joined a ‘public’ history. In terms of the themes of this paper, the idea of living in a ‘public’ world, as opposed to being shut away in a private one, is of living in a world enlarged by sympathetic participation with other people perceived as subjects.

The nature of her desire for men also changed. She met a vulnerable artist, very frightened of closeness, and felt very tenderly towards him. The jealousy that had always tormented her ceased to be concerning; she was able to be aware of his previous relationships without feeling either excited or outraged. She maintained that relationship for nearly a year, and when he ended it she was sad rather than contemptuous. She began to allow men an internal life of their own, including sexual feelings towards other women. She moved to a big city and started to participate in various courses and shared activities. I began to hear of female friends for the first time. At the age of 71, she told me that for the first time in her life she sometimes had positive feelings of happiness. She linked them in particular with the fact that she had begun to ‘go towards people’, and was no longer always fearful that good experience would be followed by terrifying retribution.

I think what we glimpse in this very brief history is the gradual re-establishment of the capacity for sympathy, which had been shocked into retreat by early traumatic experience. Ms B had been unable, perhaps for genetic reasons, to deal with early trauma by splitting, which might have led to the familiar outcome of the denial of suffering and identification with the aggressor, and perhaps to a re-enactment of a cycle of abuse. Indeed, versions of this had occurred in her earlier adult life, when her paranoid jealousy led to the abrupt abandonment of loving partners, with short-lived sadistic satisfaction. But the fact that the reaction of jealousy diminished and became manageable when she came into therapy suggests that the main component of it was fear of betrayal and abandonment rather than primarily the sadistic pleasure of punishing a bad sexual object. The fundamental posture of morality, the seeing of others not only as objects, but also as subjects, is one of the indices of a healthy capacity for sympathy; its everyday manifestation is the capacity for friendship.


Stern has spoken of ‘affect attunement’ by the mother as crucial to the affective development of the baby. He suggests that babies are acutely aware of the degree to which the mother's attunement is accurate, and may be shocked or traumatised by gross failures of accuracy (1985, 138-61). Broucek (1982) has suggested that areas in which the mother is at sea—and these might include such major departments of affective life as aggression or sexuality—will be the areas that later cause an otherwise competent child to suffer feelings of shame and internal collapse. Sympathy in what I am calling sense one is, I suggest, the elementary and involuntary capacity that makes affect attunement possible, that underlies the successful self-orientations (precursors of successful interventions) in both parents and psychoanalysts, and that allows children and patients in turn to discover their own emotional resources and to inhabit their emotional world without bewilderment. Sympathy is the capacity that makes possible the more sophisticated operation of empathy, and the more developed and specific affect of compassion that is often, confusingly, also called sympathy. I think it is also sympathy sense one, the capacity, that gives us the possibility of living in an accurately perceived emotional world where, in principle, and with the help of imagination and reflection, sound judgements of value may be made, and where a basis may be discerned for calling judgements of value sound.


Bion W (1962). Learning from experience. London: Karnac.
Bortoft H (1996). The wholeness of nature: Goethe's way of science. New York: Lindisfarne Press.
Britton R (2003). Emancipation from the superego. In: Sex, death, and the superego, p. 103-16. London: Karnac.
Broucek F (1982). Shame and its relation to early narcissistic developments.
Int. J. Psycho-Anal. 63: 369-78.
Damasio A (2000). The feeling of what happens: Body, emotion and the making of consciousness. London: Vintage.
Damasio A (2003). Looking for Spinoza: Joy, sorrow and the feeling brain. London: Heinemann.
Darwin C (1871). The descent of man. London: John Murray.
Freud A (1936). The ego and the mechanisms of defence. London: Hogarth Press.
Freud S (1900). The interpretation of dreams. SE 4-5.
Freud S (1912). Recommendations to physicians practising psycho-analysis. SE 12.

Freud S (1913). On beginning the treatment. SE 12.
Freud S (1937). Analysis terminable and interminable. SE 23.

Hinshelwood R (1989). A dictionary of Kleinian thought. London: Free Association Books.
Hume D ([1739] 2000). A treatise of human nature. Norton DF, Norton MJ, editors. Oxford: Oxford Univ. Press.
Keller EF (1983). A feeling for the organism: The life and work of Barbara McClintock. New York: WH Freeman.
Kelman H (1987). On resonant cognition.
Int. Rev. Psycho-Anal. 14: 111-23. Klein M (1926). The psychological principles of early analysis. In: Love, guilt and reparation and other works 1921-1945. London: Hogarth Press and Institute of Psychoanalysis.
Klein M (1946). Notes on some schizoid mechanisms. In: Envy and gratitude and other works 1946-1963. London: Hogarth Press and Institute of Psychoanalysis.
Kohut H (1971). The analysis of the self. New York: Int. Univ. Press. Phillipson N (1983). Adam Smith as civic moralist. In: Hont I, Ignatieff M, editors. Wealth and virtue: The shaping of political economy in the Scottish Enlightenment. Cambridge: Cambridge Univ. Press.
Pigman GW (1995). Freud and the history of empathy.
Int. J. Psycho-Anal. 76: 237-56.
Rawls J (2000). Lectures on the history of moral philosophy. Boston: Harvard Univ. Press.
Sandler J (1988). Projection, identification, projective identification. London: Karnac.

Sandler J (1993). On communication from patient to analyst: Not everything is projective identification. Int. J. Psycho-Anal. 74: 1097-107.
Smith A ([1759] 1976). The theory of moral sentiments. Raphael DD, Macfie AL, editors. Oxford: Oxford Univ. Press.
Stern D (1985). The interpersonal world of the infant: A view from psychoanalysis and developmental psychology. New York: Basic Books.

Trevarthan C, Aitken K (2001). Infant intersubjectivity: Research, theory, and clinical applications. J Child Psychol Psychiat 42: 3-48.

Please cite as:
Black, D.M. (2004). Sympathy reconfigured: Some reflections on sympathy, empathy and the discovery of values.
International Journal of Psycho-Analysis, 85:579-595